Authors - Pouyan Nahed, Mina Esmail Zahed Nojoo Kambar, Jorge Ramon Fonseca Cacho, Garam Lee, Jeffrey Cummings, Kazem Taghva Abstract - The text associated with clinical trials contains a fair amount of information that experts regularly analyse and interpret to extract in- formation. In most cases, specialists use the text to create structured data for comparing candidate trials in different aspects. As this process is tedious and time-consuming, automation is important, relying on machine learning applications in natural language processing. In this paper, we explore the feasibility of the Mechanism of Action (MOA) classification for Alzheimer's drugs by augmenting Bio BERT (Bidirectional Encoder Representations from Transformers for Biomedical Text Mining) pre-trained model with a five-layer deep neural network. This recommendation model achieves an F1 score of 97% on the dataset built based on the material from the Alzheimer's disease drug development pipeline at clinicaltrials.gov and the ALZFORUM therapeutics website. Also, we published our code on a GitHub repository 7 that is available for the public to use.
Authors - Suharti Siusin, Shazlyn Milleana Shaharudin, Nur Syarafina Mohamed, Ismail Musirin Abstract - As a result of the COVID-19 outbreak in Malaysia, educational institutions, including universities, were forced to conduct all academic activities online. If online learning continues in subsequent academic sessions, it may have a big impact on the academic performance of the student. Thus, the purpose of this study is to classify student’s academic performance by using Principal Component Analysis (PCA) before and during pandemic COVID-19. The sample consisted of 234 undergraduates from the Department of Mathematics at the Faculty of Science and Mathematics, University Pendidikan Sultan Idris (UPSI). Data collection is conducted using a questionnaire. Data was gathered from students' grades for each subject in the first semester of the session 2019/2020, which is during face-to-face learning, and each subject's grade in the second semester of the session 2019/2020, which is during online learning. The finding of the study was divided into two parts which are preliminary analysis and further analysis. According to the eigenvalues greater than one criteria (Kaisar, 1960), when using the correlation matrix in Eigen analysis, only values greater than one should be included in the analysis. As a result, the researcher decided to cut-off the dimension at the second dimension from the eigenvalues obtained by the R software because the eigenvalues for the second dimension, were greater than one which is 1.111. For further analysis, the researcher will be using the principal component one in classifying students based on their overall academic achievement and principal component two will be used in classifying in which semester did they do best in. With regard to overall academic performance, the majority of students' 51.4% were in the classification of good performance, while just 24.3% were considered excellent performance. Further, 15.7% of students are categorized as performing averagely in their academic performance, whereas 8.6% are categorized as performing below average. The results indicate that 65.7% of students performed similarly in both semesters, indicating that most students are performing in their academic performance either before or during the pandemic COVID-19, while only 14.3% of students performed in semester 1 session 2019/2020, which is during face-to-face learning, and only 20% of students performed in semester 2 session 2019/2020, which is during online learning.
Authors - Rogelio Ruzcko Tobias, Johanna Minglana, Denzel Kelsey Hernandez, Matt Ervin Mital, Rachel Edita Roxas Abstract - To ensure quality assurance, Higher Education Institutions (HEIs) implement a Quality Management System (QMS) anchored on international benchmarks like ISO 9001:2015 Standards. With the COVID-19 pandemic, quality audits have become more challenging. Also, to address the lapses due to human error and lack of technical knowledge in clause identification during audit processes, an Artificial Intelligence (AI) Enabled QMS is presented. This study successfully demonstrated how AI-enabled QMS can match audit findings in accreditation compliance reports and internal quality audit reports with the clauses of ISO 9001:2015. Audit findings corpus data gathered are within the span of the last 5 years, which serve as the data set to be employed. After data pre-processing, a long short-term memory (LSTM) deep neural network was created and trained using MATLAB. The AI model achieved a combined Classification Accuracy (CA) of 82.15% and predicted 70% of the examined audit findings in actual implementation. Further analyses illustrate how AI can be maximized in generating useful and precise and useful audit reports for HEIs to develop and implement globally competitive educational policies, programs, and standards.
Authors - Alin-Marius Stanciu Abstract - In the beginning, are presented some of the most fundamental aspects of security, like confidentiality, integrity, and availability. Then the attacker’s key motivators are mentioned. In the following parts, the focus is on the ways of designing and developing a secure software product. So, from a design point of view, the secure software development life cycle and how a security management process should take place are described in multiple phases starting from security threat and risk analysis phase to security testing phase and malware scanning. To keep track of vulnerabilities that might appear in the future, especially for 3rd party products, a security vulnerability management process should be used. Finally, in the secure development chapter are also presented some of the most common vulnerabilities and ways to assure secure code. Making source code security analysis by using static/dynamic/interactive application security testing tools with manual code reviews are some important factors in assuring secure software development.
Authors - Ernesto Rivera-Alvarado, Francisco J. Torres-Rojas Abstract - The Accelerated Processing Units (APUs) are tightly integrated heterogeneous computing platforms that contain a CPU and GPU in a single integrated circuit. As their internal computing resources use the same memory (RAM), they can efficiently share data structures. This opens up the opportunity for creating new types of specialized algorithms for the APU. In this research, we designed a specialized load distribution mechanism for the APU and used image rendering as our test load as it is a highly expensive computational task. We found out that our workload distribution mechanism provides better performance than conventional workload mechanisms for GPUs and multicore CPUs.
Authors - Hyungjun Seo, Jaechun No, Sung-soon Park Abstract - As system reliability and availability become the key driving factors in various IT services, the ability of predicting system failure is considered one of the most important aspects to avoid data and revenue loss. In this paper, we propose a machine learning based system failure prediction method, called ml-SFP (machine learning-based System Failure Prediction), which automatically enables to identify the optimal machine learning model and its hyper-parameter values generating better performance, without requiring a deep knowledge of system failure prediction. We present the performance evaluation of ml-SFP to verify its effectiveness in predicting failures.
Authors - Ahmed AlKalbani, Hamed AlBusaidi, Hepu Deng Abstract - The increasing number of cybersecurity breaches has forced organizations to apply specific strategies for ensuring business continuity and minimizing business risks. The complex socio-organizational dynamics in organizations influence the adoption of cybersecurity measures. This has created difficulties to align the implementation of cybersecurity measures with cybersecurity compliance. This study constructs a typology of cybersecurity practitioners in public organizations based on their perception of cybersecurity measures using a Q-methodology. Based on the analysis of the collected data from 14 participants in different organizations, the study reveals three distinctive typologies of cybersecurity practitioners including strategy, management and operation, and technology. Such proposed typologies can be used for developing cybersecurity mitigation strategies and understanding the prioritization of cybersecurity measures for their implementation.
Authors - Malgorzata Pankowska Abstract - Nowadays, companies are conscious of the importance of information system business analysis (BISA) and business architecture (BA) development to represent and manage Information Technology (IT) in a holistic way. Business organizations require hybrid approaches that combine different modelling techniques. This paper covers literature survey of information system analysis techniques, their applicability in university teaching and in research projects, and their acceptance by students. This paper includes a system analysis acceptance model for assessing students’ attitudes to use the system analysis modelling techniques. Furthermore, this study allows to compare BISA techniques in the aspect of their acceptance. Findings of this paper could support modelling business organization assets and activities and may encourage business analysts to simultaneously learn and apply the modelling techniques. Constantly, the most popular modelling techniques set covers business process modelling notation (BPMN), unified modeling language (UML), data flow diagram (DFD) and decision trees.
Authors - Tarek Maylaa, Feryal Windal, Halim Benhabiles, Gregory Maubon, Nathalie Maubon, Elodie Vandenhaute, Dominique Collard Abstract - In this article, we propose a hierarchical deep learning framework for the nuclei 3D reconstruction from a stack of microscopic images representing 3D cancer cell culture. The framework goes through three successive stages namely: at the slice level of the stack i) the spheroid detection and ii) their nuclei segmentation then at the stack level iii) the nuclei 3D reconstruction. For this purpose, we prepared a dataset of bright- field microscopic images acquired from 3D cultures of HeLa cells and manually annotated by the experts for both tasks (spheroids detection and nuclei segmentation). Two CNN models namely YOLOv5x and U-Net-VGG19 have been trained and validated on our dataset for the detection and the segmentation tasks respectively. For the 3D reconstruction task, the delaunay triangulation technique has been adopted by exploiting point cloud clusters that represent the segmented nuclei in the stack. Our framework offers to the biologists an efficient assisting tool for quantifying the number of spheroids and analysing the morphology of their nuclei. The conducted experiments on our generated dataset show the promising results obtained by our framework with notably an average precision of 0,892 and 0,76 on the spheroids detection and nuclei segmentation respectively. Moreover, our 3D reconstruction technique shows visually a consistent representation of nuclei in term of volumetery and shape.
Authors - Tanupriya Choudhury, Thipendra Pal Singh, Prakhar Jain, Arunachalaeshwaran V R, Tanmay Sarkar Abstract - Automation of freshness classification is an application of Deep learning and Computer Vision. Normal methods are manual labour which are time consuming and inefficient. In recent years many new technologies come into place like deep learning and computer vision, since the shape is the main freshness classification, these new technologies are proved to be very useful, since the process has been improved in the terms of both accuracy and time. This paper consists of various image processing techniques used for Radish freshness classification. Comparison among different models has been made on the bases of training and testing accuracy.
Authors - Alejandro Valencia-Arias, Lina-Marcela Cifuentes-Correa, Jefferson Quiroz-Fabra, Wilmer Londono-Celis, David Garcia-Arango, Vanesa Garcia-Pineda Abstract - This article aims to identify research trends on the relationship between conservation, education, and development in geoparks based on a bibliometric study. The methodological approach is based on an analysis of 406 scientific publications on the topic of study. These documents were obtained from a search equation used in the Scopus database. The search period was between 2003 and 2022. Among the results, it is found that there has been a growth in the subject matter in the last years, with important peaks of production in the years 2019 and 2021. In addition, authors from China and Poland appear as the most productive. Among the main results, the preference for publications in journals associated with geoparks stands out. China stands out as the most productive country with the highest impact on the topic. Among the main research trends, it was found that Geotourism and geological heritage contribute a large part to the discussions on the subject. It is integrated with sustainable development as a meeting point between education, conservation, and development.
Authors - Benita G Zulch, Mafhungo Musefuwa, Joseph Awoamim Yacim Abstract - This study aims to evaluate the socio-economic challenges that affect the inhabitants of the informal settlements in Msholozi, Mbombela, South Africa. Although many studies have been done about the challenges of Informal settlements across the world, one thing is common all informal settlements pose features that might be different due to the location. The age-long apartheid legacy that greatly influences the social and economic life of the people relative to their settlement patterns in South Africa was the main motivation for this study. To gather relevant information needed to achieve the objective of this study, household heads in Msholozi were interviewed. Findings reveal that lack of employment and poor remuneration for those with employment was paramount among the economic challenges; while the social challenges included poor facilities in school, health centre, and shopping centre among others. Interestingly, the residents of Msholozi are satisfied with their living environment. The study, however, recommends that government should be proactive in meeting the housing needs of its citizens as enshrined in the constitution to curb the proliferation of informal settlements in South Africa.
Authors - Daniel Johannes Hoffman, Derick Eben Booyens, Karl Trusler Abstract - We live in a world characterised by change. Organisations and professions wanting to survive and prosper need the capacity to adapt to change. Since the first democratic election of 1994, South Africa has experienced significant change. Organisations such as the Association of South African Quantity Surveyors (ASAQS) also adapted. Successful strategies for change require accurate information. Adding an ASAQS membership category for construction quantity surveyors (CQS) is a current issue. Quantity surveyors (QS’s) are the financial consultants of the construction industry, however, more than 17% do not work as professional consultants but are employed by contractors. These CQS’s have no professional organisation serving their interests. An ASAQS membership category fills that need, can strengthen the ASAQS membership base and expand the ASAQS’s influence in the construction industry. The ASAQS, assisted by the University of Pretoria (UP), circulated a research questionnaire to all QS’s nationally. This study compiled a profile of CQS’s from the questionnaire data to assist the ASAQS’s consideration of a CQS membership category. The profile of CQS’s confirmed them to be relatively young (almost 55% are not older than 35 years), 83% are located in provinces with major urban economic hubs, 63% are African, but less than 10% are from Coloured, Indian or other racial groups. All CQS’s have tertiary qualifications, only 25% of CQS’s are not registered with the South African Council for the Quantity Surveying Profession (SACQSP), but 34% are not ASAQS members. This profile provides much-needed information guiding the ASAQS’s decision on CQS membership.
Authors - Rashed Khamis Al Romaihi, Abdelhak Belhi, Rachid Hadjidj Abstract - Drones and Unmanned Aerial Vehicles (UAVs) are one of the cheapest and most dif-ficult to counter aerial vehicles inducing security and safety concerns in many sectors. Recent events suggest that traditional detection technologies such as radar are not effective for detecting these vehicles despite the huge investments from countries suffering from various drone attacks. Current technologies often rely on data such as images to detect and locate drones in the sky using advanced AI-based approaches. Given these challenges, the global aim of this research is to evaluate the effectiveness of various deep learning-based object detection methods on the task of visually detecting airborne drones from camera feeds. We mostly compare single-stage, two-stage, and anchor-free object recognition methods in terms of recognition performance and speed. As per the data, we use as a common benchmark the Anti-UAV dataset presented at the ICCV 2021 conference. From the results, we see that anchor-free models perform better compared to the other methods.
Authors - Michael Dietz, Matthias Werodin, Ronald Schmidt-Vollus Abstract - The concept of the digital twin (DT), which accompanies a plant or machine throughout its entire life cycle, has further reinforced the role of Virtual Commissioning (VC). Of decisive importance for the economically and technically sensible use of the DT is its creation as early as possible in the life cycle. In the field of mechanical engineering, therefore, the first digital artifacts of the DT should ideally already be created in parallel with the mechanical design phase. The first step is the kinematization of the CAD model. This paper presents a new method for the semi-automated kinematization of 3D CAD models. The method uses a verbal description of the motion sequence desired by the designer, based on DIN EN 60848 (GRAFCET) [2] and using elements from VDI Guideline 2860 (handling functions) [8]. The presented method is easy to use and self-describing to a large extent. It enables designers to define in their own domain, in parallel to their mechanical designs, the desired motion sequence of the machine they have designed in machine-readable form. The actual kinematization of the CAD model is performed automatically. At the end, the kinematized design is available in COLLADA format and can be imported via the import interfaces of common simulation tools and can therefore be used for the DT.
Authors - Moanda Diana Pholo, Yskandar Hamam, AbdelBaset Khalaf, Chunling Tu Abstract - Over 600,000 new lymphoma cases and around 280,000 lymphomarelated deaths were reported in 2020. The delayed diagnosis of lymphoma has long been a problem. However, the advent of the COVID-19 pandemic, which disrupted healthcare services worldwide, may have caused more significant delays in lymphoma diagnoses. Since lymphomas can sometimes present with symptoms like COVID-19 and can affect the lungs, there is also a risk of misdiagnosis. We collected 505 lymphoma and 180 COVID-19 case reports from ScienceDirect and applied boosting methods to classify each patient as having COVID-19 or lymphoma based on the patient’s age, gender and reported symptoms. LightGBM had the highest ROC AUC ( 0.89), meaning it best differentiates between the two diseases. Therefore, this model can be used as a screening tool to reduce the delay in lymphoma diagnosis and improve the patients’ chances of survival.
Authors - Ikram Asghar, Oche A Egaji, Mark Griffiths Abstract - Blockchain is a distributed digital ledger of cryptographically signed transactions grouped into a chain of blocks. The emergence of Blockchain has changed how information can be stored and processed securely. The research interest in Blockchain has significantly evolved as the technology has found applications beyond the finance industry. Blockchain has gone beyond the simple financial transaction application to include being used to securely share patient records, smart contracts, digital identity, and assets management. As the technology evolves, more and more industrialists and researchers explore its potential use in managing intellectual property. This paper explores the feasibility of adopting Blockchain for IP management, highlighting potential challenges and possible solutions.
Authors - Tanupriya Choudhury, Ayush Aryan, Hussain Falih Mahdi, Arunachalaeshwaran V R, Tanmay Sarkar Abstract - Precise assessment of fish freshness has a huge importance for consumers. Manual analysis takes time and sometimes lead to false assessment that could lead to various diseases. With emerging computer vision and deep learning methods, automatic system for fish freshness grading is possible. In this study, we inspected four different pre-trained CNN models for freshness classification of mourala fish into three classes. We evaluated the models on different evaluation metrics accuracy, f1-score, precision and recall. The results showed that the CNN based models can provide acceptable results and can be used for determining freshness of mourala fish.
Authors - Wania Tahir, Raja Asif Wagan, Bushra Naeem Abstract - Traffic management is a subject of improvement all over the world, may it be the emerging or developed countries. Road safety and vehicle management requires continuous improvement mechanisms to ensure the safety of public. With the widespread growth of vehicles and thus the traffic, road coercions as well as threats to life have increased, making the management of traffic as a daunting task. A mismanaged traffic administration lead to rising road accidents, overcrowding of traffic movement, merger of distinct vehicles into prohibited lanes and numerous additional coercions that could be menacing for life. In this paper, aggregated channel feature (ACF) algorithm is applied to track the vehicles. Further, the algorithm is enhanced for the traffic flow management with the use of classifications of various groups of transportation, enabling the road management infrastructure to identify the nonvehicle versus vehicle and further recognize vehicles according to the type of vehicle, size of vehicle and status of vehicle as emergency or non-emergency vehicle. This helps building up of congestion as well as avoiding coercion. The proposed implementation of ACF to the road traffic for smart cities has detected true positive rate of 80%, 89%, has detected true positive rate of 69% and 79% having nonvehicle detected with as-signed priority.
Authors - Khalda Ali, Dimah Al-Fraihat, Maram Alzaidi Abstract - Artificial intelligence (AI) ethics is a hotly debated subject. Increasingly, bothserviceproviders and users face ethical dilemmas. To ensure that the development, deployment, andusage of AI are all morally permissible, various initiatives have been launched. For themost part, it's not clear how AI-enabled businesses deal with these ethical challenges. Our paperidentified four main ethical issues of AI (i.e., “Ubiquitous surveillance, social engineering, transhumanism, machine learning issues, and metaphysical issues”). It also identifiedtwomain mitigation strategies (i.e., “Policy- level mitigation, corporate governance of AI ethics”). Many people, including those working in AI-related organisations and academia, canbenefit from these findings, but policymakers grappling with how to best handle ethical concernsgenerated by AI may find them most useful.
Authors - Bajeela Aejas, Abdelhak Belhi, Abdelaziz Bouras Abstract - Identifying and extracting information from contracts is an important task of contract analysis, which is mostly performed manually by lawyers and legal specialists. This manual analysis is a time-consuming, error-prone task. We can overcome this by automating the task of legal entity extraction using the Nat-ural Language Processing (NLP) techniques such as Named Entity Recognition (NER) and relation extraction (RE). Most NER and RE methods rely on machine learning and deep learning to identify relevant entities in natural language text. The main concern in adapting the AI methods for contract element extraction is the scarcity of annotated datasets in the legal field. Aiming at tackling this chal-lenge, we decided to prepare an annotated legal contract dataset dedicated to the NER task and RE task by manually annotating publicly available English con-tracts. This work is a part of the research aimed at automating the implementation of natural language contracts into Smart Contracts in the blockchain-based Sup-ply Chain context. This paper explains the implementation and comparison of NER models using the deep learning method (BiLSTM) and transformer-based method (BERT) for evaluating the dataset.
Authors - Komeil Raisian, Jamaiah Yahaya, Aziz Deraman, Siti Rohana Ahmad Ibrahim Abstract - Today's digitised world is concerned with software products that high-light sustainability and green compliance with social requirements and expectations. Literature investigations reveal that many studies have focused on green hardware. Nevertheless, little effort was proposed to ensure the greenness of soft-ware that complies with the specification. Green software products are vital as they can solve the problems associated with the sustainability perspective to ensure the long-term use of the software. Current literature shows that there is still a gap in the studies that focus on the social sustainability associated with software products. The green elements are still needed for further investigation in the scope of green assessment of software products. This study aims to identify the principal factor and measurements of the social sustainability requirements of a green software product. The analysis of the factor and measurements shows strong relationships between them. The results are validated through the empirical study conducted with 102 respondents.
Authors - Libero Nigro, Franco Cicirelli Abstract - This work develops a clustering method which is based on the identification of density peaks in the available data. The realization is characterized by its goal of carrying in parallel as many operations as it is possible, and to exploit current commodity many/multi core machines with shared memory. The algorithm is prototyped in Java using parallel streams and lambda expressions. The paper first describes the rationale underlying the design and implementation of the clustering method. Then the tool is practically applied to several and challenging benchmark datasets, for example admitting general not spherical clusters. The experimental results confirm efficiency and reliability of the pro-posed method. Finally, conclusions are drawn with an indication of on-going and future work.
Libero Nigro is a full professor of Computer Engineering in the Department of Informatics, Modelling, Electronics and Systems Science (DIMES) of University of Calabria, 87036 Rende (CS) Italy. He currently teaches Object Oriented Programming and Systems Programming (covering modelling... Read More →
Friday August 26, 2022 11:31am - 11:45am BST
Virtual Room BLondon, United Kingdom
Authors - I.Nigmatullina, S.Sheymardanov, M.Abramskiy Abstract - Based on recent breakthroughs in medical, pedagogical, and psychological re-search, Kazan Federal University is undertaking a project to build a scientifically based model of comprehensive assistance for children with autism spectrum disorders (ASD). The findings of recent research underline the necessity to develop and deploy novel remote support models for children with ASD and their families, as well as technology targeted at boosting educational quality. This determined the need to develop a digital platform in order to increase the success of the socialization of children by increasing the amount of time for specialists to corrective work with children, informing clients about the effectiveness of the educational process and reducing the time to collect information and prepare individual child development programs, as well as family support programs. The paper proposes an algorithm for the operation of a digital platform for an online monitoring system as well as full assistance for children with ASD who use intelligent services. The developers of the online system were guided by the ideas of inclusivity, justice, and accessibility while creating the concept. Regardless of position, income, living situations, or other considerations, all services and resources established in the online system enable children with ASD and their families, caregivers, professionals, and students with equitable access.
Authors - Jingjing Lou, Yunhan Li, Qingdong Luo, Xiyuan Wan, Pengfei Zheng Abstract - As one of the common ways of human-computer interaction, gesture control has attracted wide attention of researchers. This paper studies the problem of using the data of Leap motion gesture sensor to control the space transformation posture of six-joint collaborative robotic manipulator xArm, analyses its kinematics properties, derive the transformation matrix by coordinate transformation, and sets constraints to avoid the repetition of moving target points caused by Leap motion's typical working frequency being lower than xArm's maximum receiving frequency. Experiments show that this method can track the palm posture detected by Leap motion gesture sensor in real time, and then control xArm accurately.
Authors - Katja Saarela, Vili Huhta-Koivisto, Jukka K Nurminen Abstract - Our study focuses on the areas of social and economic sustainability in machine learning. The risk of work disability can be predicted with machine learning and using various data sources. Machine learning techniques appear to be a potential tool to support expert work and decision-making. We will present the five stakeholders of the work disability prediction - the employee, the employer, the occupational health care, the pension fund, and society. All these stakeholders should be taken into account when developing AI to support disability risk prediction. We will compare two methods with different data sources, occupational health care data and pension decision register data. There is still another stakeholder, the data scientist, who is developing the machine learning algorithms. We will present five important aspects of the data processing and algorithm design phase: non-maleficence, accountability and responsibility, transparency and explainability, justice and fairness, and respect for various human rights. These aspects need to be considered when collecting data, storing it in databases, and sharing it with others.
Authors - Janneth Chicaiza, Priscila Valdiviezo-Diaz Abstract - Recommender systems support users’ decision-making, and they are key for helping them discover resources or relevant items in an information-overloaded environment such as the web. Like other Artificial Intelligence-based applications, these systems suffer from the problem of lack of interpretability and explanation of their results. Enriching or augmenting the system output with explanations increases the users’ trustworthiness and reliability regarding the system decisions. Therefore, it is important not only to measure the performance of automatic models but also to measure the explainability of the system. In this paper, we present research related to explainable recommender systems and a demonstrative case. To illustrate how explainable recommendations can be generated, we present two scenarios based on the Tripadvisor dataset.
Authors - Ivan Nedyalkov, Georgi Georgiev, Anton Gogushev Abstract - The present paper presents a methodology for studying the generated communication traffic from power electronic devices. The individual stages of the methodology are described in details. Based on the obtained results, from the proposed methodology and analysis of these results, the most appropriate IP network for the studied power electronic device can be selected. The applicability of the proposed methodology was confirmed by an experimental study of a power electronic device – power distribution unit. The proposed methodology can be applied to any power electronic device.
Authors - Bala Moussa Biaye, Khalifa GAYE, Cherif Ahmed Tidiane Aidara Abstract - In rural areas that are often difficult to access, we have different types of installations (sanitary, hydraulic, solar, school, etc.). The remote monitoring of these installations pose a real problem. Local authorities often do not have enough resources to remotely monitor all installations. To facilitate the monitoring of these equipments, we have proposed hybrid monitoring in our works. In doing so, if the equipments are of the same type and have the same characteristics, we can easily classify from failed equipments the failure of other equipments pairs. We used the similarity calculation method to sort and classify similar failures of equipments pairs. An approach where each type of failure is represented by an attribute has been proposed. Algorithms were then used to represent the proposed models.
Authors - Amit Kumar Dwivedi, Naveen Kumar, Manik Lal Das Abstract - Strong accountability identifies the misbehaving entity’s activities and protects the honest entity from false accusations. This paper proposes a strong, accountable, distributed auditing scheme for outsourced data in public cloud setup. A proof-based solution with fine grained read-write access is used to address strong accountability and auditing of data services. A comparative analysis with similar work is carried out.
Authors - Alfonso Marino, Paolo Pariso, Michele Picariello Abstract - The aim of the present research is to assess the eGovernment spread in European Union. The ranking between European Countries has taken into account the following indicators: ICT’s Usage, Digital Public Services, Connectivity, Online Service Index, Telecommunication Infrastructure Index, Digital Skills and Gini Coefficient. The comparative study shows that the problem of the Digital Divide, despite investments in infrastructure and the possibility of accessing this information is still present. Countries that have a less efficient eGovernment system and a lower ease of response to digital technology, risk-accumulating delays even in the adoption of eGovernment services. The paradoxical result could therefore be a widening of the already existing gap instead of the exploitation of digital reduce this gap through greater efficiency of the system.
Authors - Kyle Stone, Sunday Ojo, Chunling Tu Abstract - HIV/AIDS is a major problem affecting South Africans, with 19.5% of the population within ages 15-49 living with the virus, with implications for the average life expectancy for the country. There is an online Question-and-Answer Health Forum on HIV/AIDS with a volume of questions that poses a challenge for medical doctors to provide quick and adequate answers. The vast dataset supplied by this forum presents an opportunity to employ Machine Learning– based Natural Language Processing (NLP) information extraction techniques in addressing this challenge. This paper proposes a method to explore and analyse these questions using natural language processing and a topic modelling method to extract the most relevant information for chatbots. Topic modelling is employed in using the online health forum on Health24.com, to validate the proposed method. This provides interesting insight into the forum dataset with potential uses in future research work on developing chatbot-enabled automated question answering systems.
Authors - S.Rajaprakash, S.Muthuselvan, R.Jaichandaran, M.Sai nithankumar, M.Sai rathankumar, Vishal kumar gupta Abstract - The data age information is considerably more significant in open life, since individuals' wellbeing information just concluded regardless of whether COVID'19 impacted, and furthermore connected with all medical problems information. This information used to examined and anticipate the medical problems information by Machine Learning Algorithm, and afterward anticipated information need greater security. In this way, we applied the current strategy ChaCha technique and that strategy zeroed in as it were "encryption execution" so security is less. In this paper, to apply the new ES-BR22-001 strategy and this technique has 7 stages. The 1st stage is finding the K value. The 2nd stage is applying the K value in Equation (1). The 3rd stage is find the Sk values by using Equation (1). The 4th stage is applying the Sk values in the sparse matrix. The 5th stage is sparse matrix values is converted into single line. The 6th stage is pair all the values. The final stage is all pair values will be apply in the matrix. The new ES-BR22-001 method has provide security and performance are good while compared to ChaCha method.
Authors - Ahmed Shuhaiber, Wed Alkarbi, Sara Almansoori Abstract - The increased reliance on smart technologies has caused people to con-sider smart homes. A smart home is using basic assistive devices to build a home environment that contains many technological features and home appliances that are connected and integrated. In order to adopt smart homes, it is needed that users trust this technology, and the factors that influence trust are yet to be dis-covered. Thus, the aim of this study is to gain a deeper understanding of customer trust in smart homes by empirically exploring the factors that influence customers’ trust in smart homes, understanding how those factors are intercorrelated, and the influence of power and direction of each factor. To address the research aim, an online survey is conducted to explore the perceptions of the residents of the UAE residents through a convenience sampling approach. As a result, most people believe that smart homes are reliable and competent. By collecting 158 responses and analysing them through the SEM-PLS approach, it is found that the social influence, perceived security risks, and perceived financial risks significantly impact customers’ trust in smart homes and that the social influences can significantly impact people’s perceived risks (security, privacy, and financial risks) as mediators to trust in smart homes.
Authors - Jorge Duque, Jose Joaquim Moreira, Joaquim Costa Abstract - The digital age and the interest in the areas of knowledge development in databases and Data Mining are related to information and communication technologies that contribute to the exponential growth of science data. Data Mining is a method of extracting data from systems such as Business Intelligence, Big Data e Data warehouses so that through consistent systematization is it possible to create patterns that can be moldable to answer customers. In this way, the patterns must be presented as representations of knowledge. The purpose of this article is to observe the different stages of the knowledge discovery process and analyze data using Regression and Classification models in Data Mining, to support companies in knowledge management.
Authors - Abdellatif Soussi, Angela Maria Tomasoni, Enrico Zero, Roberto Sacile Abstract - The management of risk related to the transport of dangerous goods (TDG) through its various modalities represents one of the main challenges for industrial activity, because of the catastrophic consequences and adverse effects that its occurrence may cause in case of an accident for the economy, society, and environment. In this context, the European projects "LOSE+" by land (Logistic and Safety of freight transport) and “OMD” by sea (Dangerous Goods Observatory) implement an ICT and Web-GIS-based system to continuously monitor and detect vehicles/vessels TDG and map the potential impact area generated in near-real-time in case of an accident in the continuity of land-port-sea areas. The system exploits innovative technologies to increase safety, on the one hand, on the road network by installing a network of cameras and tele-laser on the main traffic routes of the city of Genoa and Livorno (Italy), which allow the local public authority to effectively manage traffic and determine the classification of dangerous goods being transported, according to the ADR standard. On the other hand, the OMD project is developing a maritime network Web-GIS system, integrable with the previous one, to increase safety through an oil spill model. Furthermore, the platform can provide users with geo-referenced data and maps of a possible accident involving the identified dangerous goods transported in the urban, port, and sea areas. This monitoring system represents a Decision Support System for territorial governance and aims to increase the level of knowledge in case of dangerous goods accidents.
Authors - Ahmed Shuhaiber, Heyam Salem Ali Alfayadhi, May Mohammed AlAwlaqi, Maitha Abdulnasser Awadh Ali Abstract - The developed E-Bartering System enables people within a society to exchange surplus goods and items that are not needed without monetary payments and share services and experiences. This bartering system aims at achieving social and environmental goals, such as increasing community cohesion and public benefits and filling individuals' needs smartly. Thus, an improvement in communal harmony and happiness and more green and sustainable society. From an environmental point of view, the system is conscious of the importance of waste risks to the environment and the importance of reusing items and recycling them. The methodology followed for developing the system was the System Development Life Cycle, where the e-bartering website went through the phases of the planning, analysis, design, and implementation. A short survey was also shared among the community members of the United Arab Emirates as a market research tool to understand people’s willingness to accept this system and use it.
Authors - Kunihiko Takamatsu, Ikuhiro Noda, Kenya Bannaka, Katsuhiko Murakami, Yasuhiro Kozaki, Aoi Kishida, Hiroki Kabutoya, Kenichiro Mitsunari, Ryohei Adachi, Masato Omori, Yasuo Nakata Abstract - We proposed a novel concept for the sustainability of digital transformation (DX), Institutional Research (IR), and Information and Communication Technology (ICT) in university reform based on Eduinformatics in this paper. DX was proposed in 2004 by Prof. E. Stolterman and Prof. AC Fors. The concept behind DX is to digitize human behaviour and create new value. Eduinformatics is a new interdisciplinary field that includes informatics and education. Eduinformatics analyze data of students and develop novel analytical methods. In this paper, we assess the method of obtaining sustainability in DX, IR, and ICT to perform university reform. To obtain the answer of this research question, we show some practical examples. From these examples, we proposed a new matrix of feasibility and sustainability and concluded that Eduinformatics and Significant Other Groups (SOGs) are the keys to obtain sustainability in DX, IR, and ICT to improve the quality in higher education.
Authors - Jamaiah Yahaya, Aziz Deraman, Yusmadi Yah Jusoh, Sathya Nantha Kumar Abstract - During the current covid-19 pandemic, older people are encouraged to use information, communication and technology in their everyday lives and discouraged from physical interactions. In Malaysia, it shows that implementing the Movement Control Order (MCO) to control the spread of the COVID-19 pandemic saw a shift in consumer trends, prompting businesses to explore new strategies to interact with consumers. But there is no study indicating the contributions of older people in this new business environment. Several complaints were reported regarding the usability, acceptability, and suitability of these applications for the elderly. The objectives of this study: 1) to investigate issues and challenges on older citizens in current software design quality; 2) To identify current practices on software usage among the older citizens, especially in e-commerce and e-business applications; and 3) To propose a model a software design quality that meets older people requirements, limitations, and expectations. The new model helps the developers design and construct software compatible with the elderly based on a quality perspective.
Authors - Karen Faith A. Abuan, Ma. Corazon G. Fernando, Jheus Brian B. Lavilla, Ace C. Lagman, Danielle Joyce Napigkit, John Heland C. Ortega, Jewelle Mae S. Soliano Abstract - The difficulty with manually analysing and processing data is that it takes too long, prone to degradation and other problems. Government agencies cannot reduce cases due to a lack of sufficient knowledge and analysis to deter-mine the fundamental trend of RTI cases. Besides, the general public is utterly uninformed of road situations. Road cast, a Vehicle Incident Management System, is a practical approach to minimizing road incidents. This study presents the automation of the manual process of collecting and analysing incident data and transparency with the general public. The output of the study will be a web application system that will produce descriptive analytics, hotspots, and forecasting (7MA), supporting the PNP’s preparation for the projected future incident cases in the following week and determining the hazardous area in Pasig City through the hotpots. The general public will be informed of the number of road incidents in a particular barangay, entire cases in Pasig City, and other data visualizations integrated into the dashboard. Scrum Methodology was used to construct the sys-tem, and numerous users with varying responsibilities and administrator permissions were necessary to access and control the system. Alpha and Beta Testing examined the system's functionality, usability, reliability, performance, and sup-portability. The researchers used purposive sampling to survey (11) technical and (49) non-technical respondents. The researchers received a score of 4.74, which translates to "Strongly Agree." The system received positive feedback from both technical and non-technical respondents.
Authors - John Raymond Arriesgado, Justine Neil Calaguian, Alena Rose Golpeo, Ace C. Lagman, Miguel Bryan Pajarillo, Peter Carl Pardo, Maria Vicky S. Solomo, Heintjie N. Vicente Abstract - FEU Tech's College of Computer Studies hosts an annual event known as Technology Innovation on Capstone Project (TICaP), which exhibits the different capstone projects of each specialization. This event allows the students to view the different projects of the FEU Tech PBL students and recognizes the group with the best capstone project and other special awards. However, problems were encountered while managing the event resulting in disorganized planning and management. The proponents developed an event management system named TICaP HUB to solve the different problems experienced by the organizers in planning and managing the event. The team conducted a system evaluation and survey of the PBL and Non-PBL students, including FEU Tech faculties. Using the Likert Scale for the system evaluation, the web application has an over-all weighted mean of 4.72 and 4.71 for the mobile application. This outcome can be interpreted that the users strongly agreed with the FURPS evaluation of the system. With the results acquired from the respondents, the proponents conclude that using the TICaP HUB can significantly help the users efficiently manage the TICaP event and lessen the time consumed in planning and organizing for the said event.
Authors - Xuehong Wang, Chunling Tu, Pius Adewale Owolawi Abstract - Wheels are very important for a mobile robot platform. A mobile robot with Mecanum wheels can move in any direction and any degree rotation without a steering mechanism, which is playing a significant role in transportation, picking up fruits, and other services in a limited small operating space. In this paper, a mathematic model of Mecanum wheels and the controller of the platform are proposed and simulated. The mathematic model is analysed according to the relations between the angular velocity of the Mecanum wheels and the speed of the moving platform. The transfer function of the system consisted of Mecanum wheels and motors is introduced. Then, a PID controller is designed. To improve the control performance, a fuzzy neural network controller (FNNC) is designed. The simulation shows that compared to the traditional PID control, FNNC gets a shorter rising time while remains the platform in stable status, which overperforms the PID controller.
Authors - Jcyle Anne T. Balmadres, Kristine Bartolome, Roi Gerome B. Bunyi, Jeffrey Rafael B. Jacobo, Jay-ar P. Lalata5, Ace C. Lagman, Ma. Corazon Fernando-Raguro Abstract - E-commerce plays an essential role in selling products or services online because it can reach more customers than traditional retail. If the customer data is appropriately mishandled, it disrupts the business’ data organization and poor customer relationship management. The study focuses on creating an E-commerce website that efficiently handles the data and integrates a personalized hybrid recommender system. Content-based and Collaborative filtering methods were used in the recommender system to improve customer relationship management, streamline procedures, organize inventory and sales, and increase profits. Sales forecasting using ARIMA was also added to use the customer data for efficient business decisions. ISO 9126 was the software quality model used to evaluate the developed system using the software quality characteristics functionality, usability, maintainability, and efficiency. The system got an overall mean score of 4.57, which is Excellent, which means the system can perform smooth transactions from ordering up to the checkout and organized products, sales, and inventory. The integration of the recommender systems was able to give recommendations based on the customer's preferences, which enhances the user experience that may lead to an increase in sales of the business since the suggestions are tailored recommendations to the users.
Authors - Muharman Lubis, Engla Raafi, Sendy Prayogo Abstract - Data quality issues are exacerbated when information is distributed across heterogeneous siled data stores throughout the organization. The nature of this environment usually involves an architecture of values that conflicts with various formats. Even within a single database, consistent data quality is not always good unless appropriate rules are applied. Whether the information is dormant in the data warehouse or manipulated quickly by the application, the data quality is not enforced at all or is controlled by various components with inconsistent rules embedded in the code. To turn information into knowledge and harness its great value, data quality must, of course, be addressed through the application of continuous data processing, starting with proper and systematic evaluation. Therefore, using hard rules across the enterprise, not only at the database level but also at the application and process level, can help deliver services and improve customer satisfaction.
Authors - Vasileios Konstantakos, Ioannis Sofianidis, Konstantinos Kozalakis, Kostas Siozios, Stylianos Siskos, Theodore Laopoulos Abstract - Earthquake monitoring is a crucial process for human safety. Existing monitoring instruments present a high level of accuracy, along with increased size and cost because of the mechanical sensing elements they rely on. This work proposes an implementation of an earthquake monitoring device either for continuous or for non-continuous monitoring, which combines the recent advances in portable instrumentation systems along with the latest MEMS sensors. The focus is on creating a modern measuring node with improved characteristics, like high accuracy, smart and adaptive functionality, reduced energy consumption, as well as small size and cost. This system is used in a case study, targeting the evaluation and safety of school buildings, demonstrating high quality characteristics on the acquired earthquake data.
Authors - Marr Darwin T. Antonio, John Matthew B. Clemente, Ace C. Lagman, Marie Jocelle Y. Mina, Ma.Corazon Fernando-Raguro, Heintjie N. Vicente, Carl Ivan M. Yap Abstract - This capstone project aims to create a visual book with an augmented reality feature and a CMS-based website. The application uses augmented reality technology to track target images on the book and display 3D models and animations of the 30 endangered bird species in the Philippines. To elaborate, when a target image is found, the application will display the 3D model of the bird and two different animations for the user to explore. The application was developed using Vuforia and Unity; the 3D models of the endangered bird species will be modelled and animated using Autodesk Maya, Substance Painter, and ZBrush, and the design and layout of the book were created using Adobe Photoshop and Adobe InDesign. The developed system is one of the first AR books about endangered bird species in the country, attempting to disseminate information and raise awareness about the status of endangered birds using augmented reality. To prove that the application is practical and usable, the researchers surveyed 70 respondents consisting of 10 from the client's organization, 40 from the general public, ten bird lovers, and 10 I.T. professionals. Based on the survey results, the system proves to be practical and usable in disseminating information and raising awareness about the status of endangered birds. Future researchers can improve the system by adopting some features and enhancing the application so that users can still utilize the mobile application even without the visual book. The researchers also encourage future researchers to implement the application in other devices with different operating systems, such as iOS and Windows, to cater to a broader range of users.
Authors - Joseph Kaberuka, Christopher Johnson Abstract - Healthcare organizations continue to face an increasing range of cyber threats. In developing nations, considered capabilities of healthcare systems including fragmented governance, limited resources and experience; it is important to identify methods that can be beneficial in the socio-technical analysis of breaches given that at the moment the level of detail and the socio-technical focus varies between incidents – some reports don’t mention socio-technical issues at all – and without consistent methods we can’t be sure whether this was due to the fact socio-technical factors weren’t important or whether they weren’t just considered by the analyst. We address these problems by evaluating the application of two promising approaches STPA-sec (System Theoretic Process Analysis) for cyber security and a focused method for use case and misuse risk scenarios, CHASSIS (Combined Harm Analysis for Safety and Security of Information System) applied to the Rwandan healthcare organization. STPA-sec has been able to capture ranges of socio-technical risks threatening cyber security of PACS (Picture Archiving and Communication System) in this hospital by making rapid integration in a resource-constraint area. However, our results show that given a variety of sociotechnical risks in this area – interaction of fragmented components, organizational vulnerabilities, it can be hard for only a high level and abstract methodology; hence it is needed to identify other forms of breaches to the analysed system; for that reason, CHASSIS is used to support STPA-sec analysis; this provides to the local analyst an-other point of consideration in handling different socio-technical risks.
Authors - Sheikh Mufrad Hossain, Md. Mahfujur Rahman, Alistair Barros, Md Whaiduzzaman Abstract - The transition from traditional to online education is challenging and has many obstacles in various situations. Due to the Covid-19 situation, we use digital blended education from the traditional system. However, in some cases, it can harm our student’s academic performance. In this research, we aim to identify the factors that impact the student’s academic performance in online education. On the other hand, this study also finds the student CGPA (Cumulative Grade Point Average) fluctuation using machine learning classifiers. To achieve this, we survey to gather data perspective of Bangladesh private university, and this data allows us to analyze & classify using machine learning techniques such as Logistic Regression (LR), K-Nearest Neighbour (KNN), Support Vector Machine (SVM), Gaussian Naive Bayes (GNB), Decision Tree (DT), and Random Forest (RF). This study finds Random Forest (RF) outperforms the other state-of-art classifiers.
Authors - Irish C. Juanatas, Roben A. Juanatas, Jan Ceddrick L. Agbuya, Brigida Grace N. Bonan, Jon Bhonz M. Bonrostro, Stephen Ryan S. Gabutan Abstract - E-commerce shows steady growth in the marketplace since it makes the lives of people easier. Today's generation is more inclined toward convenience in purchasing goods. The COVID-19 shut down many food establishments across the Philippines, resulting in an online bakeries boom. Pastries are one of the most sought goods online, especially with the pandemic surge where physical stores are seldomly open. E-commerce with a recommender system is the trend that helps customers choose products, which helps in decision making on what to purchase. On the other hand, a mobile application counterpart could increase brand recognition and customer engagement as it is now the most effective, direct, and personalized way to deliver product information. In this study, the descriptive research method was used, with a questionnaire using the functionality, usability, reliability, performance and supportability (FURPS) serving as the instrument for testing the acceptability. The overall quality of the system was given an acceptable rating with a weighted mean of 4.07, indicating that the system's functions were well integrated, that navigation was simple, performed consistently, and that the system was accessible regardless of device.
Authors - Cleofe Alvites-Huamani, John Alexander Rojas-Montero, Janio Jadan-Guerrero, Elias Munoz-Primero Abstract - Anxiety and depression in infants is a problem that is increasing, this being a consequence of the global pandemic caused by COVID-19. The current panorama indicates that infants are possible victims of suffering from these moods at very early ages and in the worst case, it becomes the leading cause of child suicide. For this reason, it is very important to seek resources and means to mitigate the situation. The current project consisted of making a web application for the control of emotions in primary school children through modules made up of theoretical sessions that cover the management of emotions, thoughts, assertiveness, frustration, breathing and muscle relaxation. Each module in turn has an intervention session, which consists of situations that are presented to the infants with the aim of analysing the psychological impact that the theoretical sessions had, with the main purpose of improving or intervening the control of their emotions.
Authors - Anika Tabassum, Tahmidul Islam, Tajim Md. Niamat Ullah Akhund Abstract - E-health system is helping mankind in a great extent. This work proposes a smart system to manage all types of health-related records of a patient having a variety of users. Currently this type of combined system is very rare. Main purpose of the system is to gather all types of information in one database accessible at anytime from anywhere. This work resulted a cost effective, smart, secure, and maintainable web-based health history management system for doctors, patients, donors, pharmacist, and general health concern people.
Authors - Marcus Tomas T. Bautista, Aaliya Khaile B. Bolonos, Uzziel J. Camba, Laureen M. De Guzman, Marie Luvett I. Goh, Ma. Corazon F. Raguro, Liezyl S. Tolentino Abstract - E-Government systems have brought significant changes on the way public services are easily being delivered nowadays. This motivates the proponents to contribute to this innovation by developing a system named “Balangay” which makes barangay services more accessible to the constituents. This is anchored with the objectives of the study which is mainly to build a centralized database for barangays and design modules with the purpose of streamlining barangay processes. Part of it is to generate an incident heat map that will assist barangay officials in creating better programs and project developments. The study is a combination of qualitative and quantitative type of research which is reflected on the use of interview and survey questionnaires. An Iterative Water-fall Methodology was also adopted as a guide in the development process. Mean-while, the ISO 9126 Software Quality Model was used in the system evaluation wherein each response is measured using the 5-point Likert Scale and respondents were selected using purposive sampling. The result shows that out of sixteen (16) barangay constituents and sixteen (16) IT experts who participated, most of them were very satisfied with the system with a weighted mean of 4.45 and a verbal interpretation of “Very Satisfied”.
Authors - Carlos Ramos-Galarza, Janio Jadan-Guerrero, Hugo Arias-Flores, Pamela Acosta Rodas, Mónica Bolaños-Pasquel Abstract - Cyberbullying is the intentional and continuous abuse or aggression through the use of technological devices and the Internet. In cyberbullying there is a dynamic of violence between an aggressor and a victim. This type of practice is very common nowadays, since human activity on the Internet is very important, for example, through the use of social networks, learning on virtual platforms, interaction in videoconferences, online video games, etc. The great problem of cyberbullying lies in the high damage it generates in the mental health of the victim, which in many cases can lead to suicide. With this background, this article presents the conceptual design of an application for children that seeks to prevent cyberbullying in the context of children. This Smartphone application aims to help children to become aware of their behaviour and impact within the virtual spaces in which they coexist. This work seeks to reduce the problem of cyberbullying in the school context.
Authors - Nikolaos Zafeiropoulos, Argyro Mavrogiorgou, Spyridon Kleftakis, Konstantinos Mavrogiorgos, Athanasios Kiourtis, Dimosthenis Kyriazis Abstract - Stroke is the second most common cause of death globally according to the World Health Organization (WHO). Information Technology (IT), and especially Machine Learning (ML), may be beneficial and useful in many aspects of stroke management. However, the majority of the existing studies focus on the development of ML models for confronting such cases without checking the degree of confidence and reliability of the constructed models. To strengthen models’ performance, diverse metric functions have to be estimated, also finding the most important features of the underlying datasets. Thus, this paper studies whether the results from diverse ML models are true and realistic or not, based on diverse metric functions to verify that they extract efficient and reliable results. With this in mind, a plethora of models are built to predict the likelihood of stroke, referring to Support Vector Classifier, K-Nearest Neighbours, Logistic Regression, Random Forest, XGB Classifier and LGBM Classifier. All the captured results are compared based on the chosen metric functions, concluding into the most suitable and accurate model for stroke prediction.
Authors - Irish C. Juanatas, Roben A. Juanatas, Jayson Raymund D. Bermudez, Rene Christopher R. Tio, Jaden Del Rosario, Anne Paula Delos Santos, Ricardo R. Edig, Dennis Martinez Abstract - Online shopping has become one of the most prominent forms of retail for businesses. This is due to the advancement of web services, and mobile applications that have become accessible, and effective with the utilization of the Internet. Accordingly, this study aims to further scrutinize the discussed application of online shopping. Therefore, an online retail system with mobile application through Android was developed, and deployed with the purpose of managing the products, and services that are offered by the company, with the standardization of data forecasting to make accurate prediction of future trends. To standardize, and validate the attributes of the said system, a descriptive research method that used a survey instrument based on the Likert scale, and the functionality, usability, reliability, performance, and supportability (FURPS) model. The said survey instrument collected 200 responses with purposeful sampling treatment and converted into distinct inputs with the use of the weighted mean formula. The functionality, usability, and reliability were rated as acceptable, with weighted means of 4.5, 4.5, and 4.5, respectively. The performance and supportability were rated as perfectly acceptable, with weighted mean scores of 4.7, and 4.6, accordingly. The system's overall attributes were rated perfectly acceptable, with a weighted mean of 4.6, suggesting that it managed and analysed sales, services, and inventory data.
Authors - Janio Jadan-Guerrero, Monica Mendoza, Cleofe Alvites-Huamani Abstract - Technology has the potential to redefine museums, opening the opportunity of new ways to connect with visitors. In the pandemic, museums had to close their doors and in-person visits were cancelled. In reaction to this reality, many museums rushed to generate digital tours of their works and collections. The creation and development of these visits was born from the need to continue with virtual visits to continue to bring the knowledge that these institutions house closer. Today a possible approach is to create hybrid experiences that utilize digital technology capable of enriching a physical visit to museums. In this context, this paper addresses three experiences developed in Ecuador: 1) The Spiral of Memory proposal developed at the National Museum of Ecuador (MUNA), 2) The Animals of Yesterday and Today proposal developed at the Smithsonian Virtual Museum and 3) Interactive Museums as Pedagogical Mediators, digital educational resources in virtual learning environments. The conclusion of these experiences led to the conclusion that museums are spaces of interaction that contain an infinity of resources that allow children to create their own stories and compositions from the observation of real facts and situations, presented through the creation of physical or virtual experiences in these cultural spaces.
Authors - Visit Hirankitti Abstract - The future of factory automation is certainly Industry 4.0, where SCADA software is its essential element. A more advanced SCADA software, namely “Intelligent SCADA,” is needed to be developed to support Industry 4.0. In this paper we shall investigate an intelligent SCADA software by pro-posing an object-oriented agent framework for it. The proposed framework is modeled entirely using the object-oriented approach; with its merit we can rep-resent and reason about machines’ state and change, and we can also model and reason about object language and meta language of what we call “a cyber-physical system.” Moreover, we gain a great deal of benefit from object-oriented concepts for this intelligent SCADA, such as inheritance, polymorphism, persistence, object-oriented design patterns and so on. We shall argue that our approach captures the notion of a cyber-physical sys-tem and a digital twin naturally as well as embraces logic fundamentally. The framework was developed, as a web-based SCADA application, using Python on the server side, and JavaScript together with HTML on the client side.
Authors - U Ananthanagu, Pooja Agarwal Abstract - In recent couple of years, social media platform has attained a tremendous growth in sharing information and become an integral part of everyday life. Such real time information portrays one’s daily activities by exposing their private space, behaviour aspects and emotions. With the wide range of availability of personal information on social media, it plays a prominent role on the process of individual’s mental illness detection. Although, in the last few decades mental illness rates create more attention and also many cases are not recognized as now. Indications related to the mental illness are visible on web forums, Facebook and Twitter henceforth computerized techniques are in demand in earlier detection of mental illnesses and depression. In this case, automated analysis of mental illness detection on social media may provide early accurate detection process. Studies have so far explored how the use of social networking sites is linked to mental illness in users or attempts to diagnose mental illness by analysing user-generated content. The review focuses on the confines of machine learning and deep learning techniques on the diagnosis of mental illness of the publications regulated from 2017 to 2021. Amidst current depression detection algorithm, we spotlight various transfer learning categories and scrutinize the issues. Finally, the presented work providing the research guidelines to the upcoming research works for the detection of mental illness based on transfer learning deep learning. The main motive of the manuscript focuses on future prospects of transfer leering based mental illness detection on digital media stream.
Authors - Lizeth Rodríguez, Adrian Muros, Oriol Paris Abstract - In relation to embodied carbon, it is important to recognize that, building are a significant material bank, as they constitute a repository of high-carbon resources for many decades, according to Level(s) indicator 1.2 Thus, it is important to study designs that facilitate future reuse and recycling at the end of the building's useful life. Until now, the practice of evaluating reversibility in building materials to establish the potential to avoid impacts, has been at an experimental level and for research purposes to verify the percentage of recoverability of materials at the end of their life cycle. This work present Life-cycle assessment of three buildings representing the design criteria efficiency, both in energy consumption and the optimal use of materials is carried out, under the "Cradle to Cradle" approach, considering manufacturing as the cradle and the reuse scenario as the second cradle, evaluating the circularity of the components classified as "critical elements" in indicator 1.2. The reduction in the carbon footprint in the life cycle of a building over a period of 50 years is achieved thanks to the benefits beyond the system, especially in two relevant aspects: 1. the independence of the use of fossil resource, which is the case of NZEB buildings. 2. The closure of material cycles, through durability, reversibility and recyclability
Authors - Pornpimol Chaiwuttisak Abstract - This research aimed to predict volume workloads in debt collections and compare the accuracy of forecasting methods. Three forecasting methods were considered in the study: the SARIMA model, the Random Forest model, and Long Short-Term Memory. The data used in this study was a time series of a daily debt collection volume workload due 1, 5, 10, 15, 20, and 25 in the used car loan corporation. They were divided into 2 data sets. The first data set containing the past data from June 2020 to October 2020 was used for selecting the most suitable model, and the second data set containing the past data from November 2020 to December 2020 was used for comparing the prediction accuracy of each forecasting model in terms of Mean Absolute Percentage Error (MAPE). Consequently, the lowest MAPE achieved by LSTM for different due dates was 6.79%, 6.24%, 7.08%, 10.88%, 12.45%, 10.17% respectively, while the MAPE achieved by Random Forest was 11.72%, 11.27%, 8.94%, 11.92%, 13.91%, 16.67% respectively, and the MAPE achieved by SARIMA was 23.35%, 32.87%, 32.58%, 30.07%, 33.93%, 27.31% respectively. It indicates that the LSTM model was the most accurate to forecast the daily debt collection volume workload of the field debt collector in advance of the future. The predicted workload can be used as a piece of supporting information for adequate workforce planning of the corporation’s used car loan.
Authors - Thanh D. Nguyen, Uyen U. T. Banh, Tuan M. Nguyen, Tuan T. Nguyen Abstract - E–channels are fast replacing traditional channels as a means of shopping and consumption. E–service is the fusion of two trends: the shift from the commodity economy to services and the expansion of the information economy and the electronic networks. E–service quality (e–SQ) is very significant in the electronic environment. Thus, the studies about e–SQ are vital and meaningful. This study approaches the related concepts to e–services and e–SQ. Besides, this study reviews the related models and scales to e–services and e–SQ. Research also indicates the challenges and research trends related to e–SQ. This research exerts scientific literature synthesis, prioritizing published articles from 2000 to 2020, and articles have been published in the scientific journal with international standards. In addition, this study also analyses, compares, reviews and evaluates the related issues to e–SQ.
Authors - Alexander Gretchenko, Liudmila Goncharenko, Sergey Sybachin Abstract - This article summarizes the history of the discovery of N.D. Kondratiev of large cycles of economic conditions, as well as the creation and justification of the theory of innovation-cyclical economic development of Kondratiev-Schumpeter. The main conclusion in this article is that in general terms today it can be argued that the Kondratiev-Schumpeter theory is sufficiently substantiated. Further, the possibility of making a forecast of the development of the economic situation in the direction of applying this theory in practice, which demonstrate its effectiveness, is considered.
Authors - Miglena Temelkova Abstract - The present article studies and analyzes the nature and specific characteristics of the cyber-physical production systems, and on that basis it generates a model for their organization. Following research and critical analysis of the definitions of the concepts “production system” and “cyber-physical system”, there is specified the definition of the concept “cyber-physical production sys- tem”. There is an analysis of the structure of a cyber-physical production sys-tem made in the course of research, which is the scientific-and-methodological basis of the model for the organization of the cyber-physical production systems generated with the present article for the first time in research literature. This is the first scientific research, which studies and analyzes the purpose, functions and content of the structural organizational elements of a cyber-physical production system, as well as the relational dependences and mutual determination between them. The study also points out the organizational, management, soft-ware and programming methods, tools and approaches, which ensure each of the structural elements of a cyber-physical production system. Outlined are problems unsolved by science, which are related to the topic, as well as the debating is-sues and the results, which are product of the present research work. Some conclusions on the topic are synthesized.
Authors - Boris Kontsevoi, Sergey Kizyan Abstract - The paper explores the Predictive Software Engineering (PSE) framework and covers the seven principles listed therein, which work together to ensure transparency, controllability, and predictability. The framework provides a reliable, systematic approach for delivering development services. Thus, the article aims to help readers transform their custom software products into successful, high-quality business solutions. Throughout the article, the authors investigate how PSE principles allow be-spoke software development companies to offer transparent products and services that fit budget constraints. The paper examines each principle, one by one, providing strategies, best practices, methodologies, and relevant KPIs. The PSE principles were created taking into account 27+ years of software development experience and global engineering best practices.
Ceo & Founder of Web3 Digital Marketing LLC., Intetics Digital Marketing Director. IVLP`2018 alumna of the U.S. Department of State program on Women and Entrepreneurship, TOP 100 Women of the Future in Emerging Tech.Expertise: Startups scale up more than 8 years Digital Marketing... Read More →
Friday August 26, 2022 4:31pm - 4:45pm BST
Virtual Room ALondon, United Kingdom
Authors - Tate Seyler, Denis Ulybyshev Abstract - It is vital for cyber-physical systems to quickly recover after emergencies that might be caused, for example, by natural disasters or cyber attacks. In this paper, we propose a secure field diagnostics methodology for monitoring cyber-physical systems. Our solution relies on near-field communication technology and a mobile application that incorporates biometric fingerprinting for its fast and secure role-based authentication. For data protection, we propose an encryption schema with on-the-fly key generation, using a biometric secret. Our solution also allows operators to read and write diagnostics messages for field devices and sensors, which makes the maintenance of cyber-physical systems easier and reduces the probability of failures.
Authors - Valerii L. Muzykant, Victor V. Barabash, Olga V. Shlykova, Kirill P. Barsukov, Munadhil Abdul Muqsith, Radita Gora Tayibnapis Abstract - Smart cities such as Moscow and Jakarta manifest rapidly updated de-sign technologies, means of communication, new values, and communicative meanings. Moscow was one of the first to receive a certificate of successful implementation of international key performance indicators (KPIs) for smart, sustainable cities. About 330 online services made it possible to realize a number of valuable for the citizens’ projects base on people’s active social position like «Our City» as well as «Active Citizen» based on newly created government's online platform. Likewise, in agriculture in Indonesia, which still relies on human labour for processing, the technology has not been optimized for irrigation in rice fields or large plantations. However, the use of Drone system technology (Drone technology) is done by PT. Perkebunan Nusantara (PTPN) carries out irrigation with automation technology. In Moscow almost 20,000 access points to free un-limited Internet have been successfully started. One of the innovative The IT product is Ruli (Drive) innovative service project as one of the Moscow Government's 18 state programs aimed at further developing infrastructure and computation in a smart city as e-government contemporary competence. In 2021, Indonesian Gojek made a big breakthrough by merging with one of the big e-commerce companies, Tokopedia amounted to 135.1 million users. The article analyses the approach and results in the field of ICT (Information and Communication Technologies) in the governance process for providing access to information and services to the citizens, thus encouraging them to participate in the administrative smart city process.
Authors - Zlatinka Kovacheva, Ina Naydenova, Kalinka Kaloyanova, Stoyan Poryazov Abstract - This article presents a multidimensional model of WSN data quality that links the types of errors with the methods for their detection and correction. The designed model aims to support the wireless sensor network management. It provides a possibility to monitor the status of the network regarding the sensor data quality indicators and to predict the future development of the network.
Authors - Gunet Eroglu Abstract - Auto Train Brain is a mobile app that improves reading speed and reading comprehension in dyslexia. The efficacy of Auto Train Brain was proven with a clinical trial. We have analysed the long-term training effects of the Auto Train Brain on dyslexic children. We have collected QEEG data from 14 channels from 21 dyslexic children for 100 sessions and calculated the Sample Entropy in the gamma bands for the left posterior brain (T7, P7, and O1). Although the gamma band values fluctuate and no permanent increase in the gam-ma band values is detected after Auto Train Brain training at T7, P7, and O1, the variance of gamma-band sample entropy increases as the neurofeedback session number increases. We have concluded that the Auto Train Brain in-creases the flexibility of the left brain in dyslexia.
Authors - Martha Torres, Virgilio Gonzalez Abstract - Considering the complexity and multiple alternatives for technology decisions in Wireless Sensor Networks (WSNs), a multicriteria selection method (MCDM) is an appropriate approach for choosing the best option in technical projects. At the present, pure quantitative decision methods have been developed based on the customer requirements and opinions provided by experts from various fields. In this context, technical problems and additional costs can affect the implementation and operation stages. In order to prevent future difficulties and obtain a more accurate technology selection, a new method is being developed to involve qualitative and quantitative parameters taken from real scenarios and technical literature review, and optimized with a neural network de-sign. This paper provides a detailed description of this methodology involving the Analytic Hierarchy Process (AHP) and real scenario simulations used for Neural network training, as well as, risks and economical analysis for the final decision.
Authors - Mezui Eya’a Guy Lysmos, Mostafa Hanoune Abstract - We present in this article a hybrid and advanced information system in an intelligent gateway for health. Indeed, our intelligent gateway has a set of modules that allows: Firstly, ensure good communication with applications, peripherals and external devices through the use of secure communications protocols; In a second step, to properly manage the information processing system coming from the sensors thanks to advanced data processing methods (centralized, random and dynamic) and finally in a third step, to optimize not only the bandwidth of the network but also the local and remote data storage space through the use of modern and personalized data compression mechanisms (predictive compression, compression by sampling and compression by scheduling). During this approach we used the global variables to determine the limits of the volumes of data stored during a determined period to control the stop and the resumption of data transfer. Finally, a comparative study of information systems in a gateway has been highlighted to highlight the performance and efficiency between the traditional information system and that of the modern one (ours).
Authors - David Alvarado, Margarita Zambrano, Cesar Villacis, Fernando Galarraga Abstract - The objective of this study is to stimulate the conceptual, procedural and attitudinal development of high school students who take subjects considered as complex, such as Programming, applying the Problem-Based Learning for Computational Thinking Development Model (CTPBL). The agile Extreme Pro-gramming (XP) methodology was applied for the design and development of the online course template from the point of view of Software Engineering. In addi-tion, the ADDIE model was applied for the instructional design of the online course, from the educational point of view. As a proof of concept, a basic course on Structured Programming in the C/C++ Language was implemented, using a template developed on the MEAN Stack Web, which was uploaded to Google’s virtual classroom platform. For the evaluation of the online course, an evaluation matrix was applied from the student’s perspective and a matrix of usability met-rics for online courses was applied too. The results show the validation of the theoretical and practical procedure applied to the course with the CTPBL model from the technical point of view, for which the two Cearreta evaluation matrices were applied. The first matrix of Cearreta allows to analyze the performance of the students regarding the development of abilities and skills in the resolution of problems and development of projects focused on the area of Programming as part of CT. The second matrix of Cearreta allows to analyze the mastery of knowledge by the teachers who teach Programming subjects according to the ac-ademic curriculum that is completed in high school.
Authors - Miglena Temelkova Abstract - Under the conditions of Industry 4.0, the restructuring of the real-life production systems into cyber-physical production systems requires the rationalization and systematization of the specialized knowledge into a concept of interrelated and interdependent notions, definitions, principles, functional dependencies, structural elements. The analytical arrangement in a conceptual system of fundamental knowledge, tools and practices in respect to the process of restructuring the production systems into cyber-physical ones contributes to the more comprehensive understanding of that process and to focusing scientific thought on specific problems and scientific bottlenecks. The present study generates for the first time in the scientific literature a concept of the restructuring of traditional production systems in innovative cyber-physical production systems. The analysis examines the specific boundaries, elements and scope of the restructuring process and defines the main relational dependencies between the constituent functional planes of production systems in the course of their transformation and upgrading to cyber-physical production systems.
Authors - Christos Goumopoulos, Christos Karapapas Abstract - Exergaming technologies have rejuvenated the perspective of using game environments as facilitators of exercising adherence for health promotion. The entertaining character of exergames and the high accessibility of modern pervasive computing technologies have enabled a more sustained embracement of this niche technology, even by elderly adults. In spite of this view, personalization remains an open research issue with anemic solutions even though it rep-resents a critical aspect to enhance exergaming usefulness. This paper explores notions for building an adaptive exergame platform that employs off-the-shelf body tracking sensors to collect movement data during gameplay and creates real-time game adaptations based on performance metrics and machine learning models. The adaptation layer of the platform provides the system intelligence for data analysis capable of adapting the gameplay in terms of adjustments of game parameters to match the skills of each player, thus resulting in a personalized game experience. The importance of an adaptive platform in the exergaming domain of the elderly is discussed and methods for its implementation are suggested.
Authors - Carlos Ramos-Galarza, Maria Judith Lopez-Cardenas, Jorge Cruz-Cardenas Abstract - Eating disorders are a wide range of abnormal behaviors related to food and the vision of the body, they are diseases that can cause permanent damage to the human being until death. It has been seen that there is a close relationship in the impact of social networks and the use of various technologies that have led new generations to shape their perspective of the body and their food in a different and complicated way. Technology becomes the link that people use to relate to the world and allows them to create and transform themselves around body stereotypes that lead them to modify the quantity and quality of their food. On the other hand, technology also provides the human being with an addictive and possibly sedentary part, which makes the person stop being active, having as a consequence anomalies in their diet such as weight gain. This article aims to consider the role of technology in Eating Disorders.
Authors - Kamal Upreti, Prateet Mishra, Manish Maheshwari, Prashant Vats, Aisha Dhankar, Reenu Batra, Saneh Lata Yadav, Tanvi Chawla, Jayant Bhardwaj Abstract - Pneumonia is an infectious ailment which affects the respiratory sys-tem and is caused by microbes that infect the human by affecting their lung air sacs and fill them with fluid. Using Chest X-ray is the most popular approach for detecting pneumonia, and the results must be calculated and examined by medi-cal experts. The inconvenient way of detecting pneumonia results in a loss of human life owing to incorrect approach and treatment. With the advancement of computer and information technology, developing an autonomous system that can help in the diagnosis and prevention of pneumonia is now viable, especially if the patient is in a place where the medical facilities and expert doctors are limited. In order to solve this challenge, this project uses deep learning technol-ogies. The Neural Network was customized to accomplish the difficult task of detecting diseases such as pneumonia in order to aid medical experts in their di-agnosis and treatment options. CNN applies a filter to a photo in order to extract data from a chest X-ray. The above-mentioned goal was achieved using Python programming language, as well as deep learning and neural network methodolo-gies.
Authors - Estefany Yoselyn Torres Tineo, John Milthon Cordova Vasquez, Mario Roberto Huerta Mendez, Omar Alfonso Santos Pacheco, Mario Ninaquispe Soto, Yasmina Riega-Viru Abstract - This research work develops an analysis based on the results of the simulation of the production flow of a PET plastic bottle factory in San Juan de Lurigancho, Lima - Peru, using data collection techniques through the review of records and direct observation. Likewise, data processing techniques were used through the application of the statistical software Stat Fit, which was processed and simulated by the ProModel software. The objective is to generate a proposal to improve the production flow of the companies in this area, based on the results obtained in the case study model. Among the results, it was found that the reorganization of the workers' functions is more efficient, and by improving the percentage of machine utilization, the incidence of bottlenecks is reduced. The results were validated by applying the statistical test called Tests of Means. Finally, the findings showed that it is possible to improve time in the production flow. It was also found that it is not necessary to hire or dismiss someone for improvement, only a better organization of the points involved in the process was needed.
Authors - Jawad Awawda, Isam Ishaq Abstract - Agriculture is one of the main sources of income in the economy. One of the key components in an agriculture system is its irrigation system. The efficiency and practicality of the irrigation system will have a direct impact on crops yield. One of the main challenges for an irrigation system is water scarcity. Another challenge is time, based on the used irrigation system the task of watering the field could be labour intensive and time-consuming. To maximize crop yield, farmers must practice precision agriculture, and that is another challenge. A smart irrigation system is part of the solution to practice precision agriculture and thereby getting the most from what resources are available. This paper presents a smart irrigation system that addresses the three mentioned challenges to pave the way for a solution to precision agriculture. The system will measure the water tank levels, soil moisture, humidity, temperature, rain levels and will fetch the weather forecast for temperature and rain levels and using its algorithm it will decide when to start the irrigation process and for how long. The system will provide the farmers with a web portal that contains the system status that’s includes water tank level, water pump status, weather fore-cast, sensors measurements.
Authors - Renzo Segura Zavalaga, Mitshell Callacna Vera, Gabriel Guevara Alarcon, Erick Gutierrez Fernandez, Mario Ninaquispe Soto, Yasmina Riega-Viru Abstract - In this research the analysis and study of a waiting line for customer service in a swimming academy in Lima, Peru is addressed. The waiting system was modelled in the ProModel simulation software to determine the efficiency indicators of this waiting line, and to know if this process is being adequately managed or not, given certain service standards. It was determined that the waiting line is not being adequately managed and an improvement to reduce the waiting time in such a way that it complies with the policies established in the organization was proposed.
Authors - Ivaylo Chenchev Abstract - Every blockchain block header contains an attribute called Merkleroot. It is the calculated hash from all included transactions into a particular block. For example, in bitcoin, the function used for its calculation is double SHA-256. Few questions are rising: What if another hash algorithm is used? What if another hash algorithm with a different digest size is used? What if a combination of a few different hash algorithms is used? How about reflecting all those changes to the performance compared to the classical usage of the double SHA-256 function. This paper highlights, summarizes, and extends a part of the author's Ph.D. thesis research for blockchain security improvement and, at the same time, speed-up the calculation of block headers. The proposed method divides the calculation of the Merkle-root block-header attribute into four groups. The measures in each group are done with just one hash function. While in the different groups are used different hash-functions. The paper's focus is around three of the possible options for arranging the hash functions in the groups. All calculated results from the three types of group arrangements are compared to the usage of the classical double SHA-256 function. The results show that for two group arrangements, the calculation speed is improved, regardless of the used compute infrastructure.
Authors - Zineb Slaoui, Sanaa Faquir, Khalid Alaoui Zidani, Ali Yahyaouy Abstract - Worldwide warming could be a critical concern that raises a require for cleaner energy generation with the aim of reducing carbon emission impacting negatively on the global climate. As it is entirely produced from renewable sources, green hydrogen is considered the clean energy of tomorrow, it must play a key role in the energy transition the production of green hydrogen is strongly affected by the intermittent behaviour of renewable generators. There are several methods used to produce this type of energy that can be used in different sectors. Morocco is one of the countries interested by this type of energy and started projects to produce green hydrogen and used it especially in the industrial sector. This paper proposes a comparative study between the different methods of producing green hydrogen and proposes a new approach to manage the flow of energy in the industrial sector. The produced hydrogen provided by renewable sources can be used either when batteries are discharged to their minimum with no load satisfaction or when the storage batteries are charged in their maximum and the surplus of energy produced by the renewable sources is lost. The amount of oxygen derived can also be stored or sold in the industrial market.
Authors - Tachanat Akarajaka, Komgrit Leksakul, Chichana Suedumrong, Nivit Charoenchai Abstract - Cardiac Arrhythmia has grouped into cardiovascular disease, and it occurs when the electrical signals that coordinate the heart's beats didn't work properly. Moreover, it can cause the body to have abnormalities ranging from mild to death. Cardiac Arrhythmia is among the leading death rate in the world. It can occur from several reasons such as sleep deprivation, eating fatty foods, and lack of exercise. From the structure of the heart system, it was found that there are 4 functional electrical pathways e.g., SA Node, AV Node, Bundle Branches, and Purkinje fiber. These electrical signals can be read by EKG. In this experiment, 10,000 images of EKG from PhysioBank ATM were used and divided into 4 classes as abnormal SA+ AV Node, abnormal Bundle Branches, abnormal Purkinje fiber, and Normal condition to train and compare the result of using two CNN models: transfer learning MobileNetV2 model and transfer learning Xception model. Then, these models were used to detect Cardiac Arrhythmia. As a result, the transfer learning MobileNetV2 model has an accuracy of 98.58%. Besides, the transfer learning Xception model coved an accuracy of 94.51%. It can be concluded that the transfer learning MobileNetV2 has higher accuracy than the transfer learning Xception at 4.34%.
Authors - Yarmak Anastasia Abstract - In providing secure group-oriented communication and data access control, one important task is to establish a shared key between group members. To build such protocols, various mathematical apparatus can be used. Most modern group key establishment schemes are a generalization of the Diffie-Hellman key agreement protocol. This paper presents a group key agreement protocol based on assumptions relating to isogeny of supersingular elliptic curves. The properties of isogeny graphs, as well as the abundance of hard assumptions, make it possible to build flexible protocols. The proposed scheme is decentralized and implies the presence of a trusted party (group manager). Establishing a shared key is carried out in 2 rounds, one of which is aimed at con-firming the identity of the group users. Analysis of the proposed protocol security are given. In addition, performance characteristics show that there are restrictions on the size of groups due to the need to calculate the isogenies of elliptic curves.
Authors - Athraa Juhi Jani Abstract - A nanomachine has the ability to proceed simple functions, as an example: computing, sensing, actuating and/or data sorting. However, if there are more nanomachines, it is possible to achieve more complex functions through communication and information sharing. Thus, a nanonetwork would be formed through the communicating and sharing information among few nanomachines. We consider in this paper a model of N nanomachines which are located in two-dimensional environment, communicating between each other via diffusion to form a nanonetwork. The model aims to estimate the number of nanomachines in the nanonetwork, i.e. N. This paper presents two algorithms for the estimation. The first proposed algorithm is based on quorum sensing mechanism. Where, quorum sensing represents a biological mechanism which is utilized to enable the synchronization of bacteria population. While the second algorithm; we employed a counting algorithm in beeping model to estimate our nanonetwork size. Beeping model is a group of nodes that are communicating through multiple-access channel, where, the communication concept of beeping and listing to the channel is similar to the one in diffusion based molecular communication.
Authors - Dewar Rico-Bautista, Efren Romero-Riano, Yurley Medina-Cardenas, Claudia Jazmin Galeano-Barrera, Fabian R. Cuesta-Quintero, Edwin Barrientos-Avendano, Luis A. Coronel-Rojas, Yesenia Areniz-Arevalo, Jose Swaminathan, Nolfer Rico-Bautista Abstract - The smart university concept seeks to improve the quality of life by applying information technologies in a comprehensive, intensive, and sustainable way. It is an emerging concept, where little attention has been paid so far to the quantitative description of its evolution. The main objective of this article is to present the results of a bibliometric analysis around the emerging concept of smart university. It includes citation analysis, keyword networks, journal co-citation networks, references, and most cited authors. For this paper, a computer tool was used as a support to identify and visualize the intellectual structure of the smart university concept and its relationships retrieved from the Dimensional platform.
Authors - Md. Masrul Sakib, Meem Karim, Aftab Miraj Swachchha, Maheen Islam Abstract - Corona virus disease (COVID-19) is one of the deadliest scourge man-kind have ever seen. It's a highly infectious influenza virus which may transmit from one person to another without causing any symptoms. In compliance with WHO (World Health Organization) data, the 2019 corona virus (COVID-19) was first found in China and has spread swiftly to individuals in other countries, with an estimated total of 349,641,119 cases (till 25 January) globally. As counter measures to this condition, screening afflicted people is mandatory which re-quires time and is also cost-effective. Radiological scanning is a plausible measure for achieving this aim. In this case chest X-Ray is the most at hand and cost-effective alternative. We present a Deep CNN (Convolutional Neural Network)-oriented method for perceiving COVID-19(+ve) infected people by analyzing chest X-Ray images in this research. To identify corona virus and patients infected by analyzing chest X-ray radiographs, four pre-trained CNN (convolutional neural network) models (AlexNet, VGG16, InceptionV3, EfficientNetB4) were suggested in this study. Among these models EfficientNet B4 gives us the highest accuracy to detect if the person has COVID-19 or not.
Authors - Jesus Gerardo Torres-Vega, Juan C. Cuevas-Tello, Cesar Puente, Jose Nunez-Varela, Carlos Soubervielle-Montalvo Abstract - Handicapped people represent fifteen percent of the world population. Autonomy is one of the key things that disabled people seek to have. Intelligent wheelchairs could contribute to people’s autonomy by removing assistance needs and performing tasks such as autonomous navigation, real-time object detection and obstacle avoidance. The present work proposes a new intelligent electric wheelchair architecture, and then focuses only on the computer vision module. To test this module, a new dataset was created with twenty different classes, using three different vendor cameras at the same location with the same illumination conditions. We employ YOLOv4, a real-time object detector based on CNNs, installed into an embedded system, the LattePanda Alpha 864s minicomputer. Mean Average Precision was the metric used to evaluate the performance, in terms of localization and classification of objects. The iDS camera, an RGB colour format and high-resolution images, demonstrated to have the best performance.
Authors - Ivaylo Chenchev Abstract - The blockchain technology attracted great scientific and applied interest in the last six-seven years. It successfully applies in different areas like fintech, smart cities, IoT, healthcare, banking, retail, oil industries, and many others. Regardless of the blockchain type (permissioned or permissionless), the data that must be stored in the blockchain network. The algorithm, which is used to “decide” whether to approve (confirm) or not storing the data in the network is the consensus algorithm. It is very important component of the proper blockchain functioning. The author is pursuing two goals with this research. The first one is to make a thorough retrieval of articles from different publishers, based on few keywords, and check if there are any specific trends of research interests in the recent years around these keywords. The second goal is to identify some classifications criteria – like types (public, private, hybrid, federated), like layers (two layers, multi-layer), etc., and finally, to classify the existing Distributed Ledger Technology (DLT) consensus algorithms with focus on blockchain, based on the architecture and on the access type.
Dr Roshan Ragel has been a professor since October 2017. He has been a Professional Member of the IEEE and IEEE Computer Society since 2005 and a Senior Member since 2014. Prof Ragel has co-authored over 200 peer-reviewed articles on the Internet of Things, Wearable Computing, Bioinformatics... Read More →
Saturday August 27, 2022 11:01am - 11:02am BST
Virtual Room ALondon, United Kingdom
Authors - Djiwa N’tela OGA, Pelagie HOUNGUE, Cheikh SARR Abstract - Nowadays, the Internet of Things (IoT) is applied in several domains such as health, domotic, transport, agriculture and so on. This research work proposes a knowledge graph analysis of hospital Information Systems applied to IoT. We used Dimensions base collection as a data source combined with VOSviewer software to establish different possible joins in order to bring out some relevant results. Thus, bibliometric methods for the analysis of 4,743 articles published in the field of IoT security and data protection for hospital Information Systems have been provided. The results show that the field of IoT-based health research born from 2014 continues to grow. This study provides important knowledge support for further research. In the coming days, we will focus on the state of the art of IoT security and data protection for hospital information systems for remote and real-time patient monitoring.
Authors - Elena Shkorkina Abstract - An important condition for the functioning of systems such as mobile, vehicular, or flying ad-hoc networks (MANET, VANET, VANET), the Internet of Things (IoT) and cyber-physical systems as a whole has recently become the real time data processing and analysis. To ensure this aspect, edge computing have been invented and are being developed. The article describes the edge architecture of the IoT network and considers various models of device interaction within it. An authentication protocol with key agreement for the Internet of Things (IoT) network based on the edge computing architecture is proposed. The use of the protocol allows to authenticate intelligent electronic devices on another IoT device through the edge server (ES) and provides ensuring high resistance to various attacks. Computational load on resource-constrained devices is reduced by delegating computation to the ES. The flexibility of the protocol in terms of cryptographic algorithms allows to select them depending on the hardware characteristics of the IoT devices. An interaction phase for post-authentication management is also described.
Authors - Marion Adebiyi, Miracle Nmesomachi Enwere, Timothy Adeliyi, Abiodun Okunola, Ayodele Adebiyi Abstract - Genetic mutations give rise to a quasispecies of drug/vaccine-resistant and virulent organisms. These organisms are classified as strains or variants depending on the extent of their phenotypic manifestation. Thus, there is a thin dichotomy between SARS-CoV-2 strains and their associated variants. This paper sought to comprehensively review the successes achieved in the classification of SARS-CoV-2 strains based on genomic sequences (GS) using deep learning architectures, thereby stimulating further research on the variants identified recently. Selective screening and analysis of research articles centered on deep learning architectures employed for SARS-CoV-2 detection based on GS information were carried out. This incorporated the use of relevant key/search terms and logical/Boolean operators to scan through the Scopus repository. To provide a foundation for future investigations on the classification of SARS-CoV-2 strains, meticulous analysis of the three key aspects, such as abstract, methodology, and conclusion, was implemented. Despite the high level of intra-species similarity, this article presents new studies that use deep learning technology to detect SARS-CoV-2 strains on the premise of the primary sequence of nucleotides in their genome. Manually searching through specific genes for mutations to identify variants after sequencing can be very laborious. This is where the use of computational acumen comes into play. Deep learning, an offshoot of machine learning, has been utilized in various literature to tackle such problems. Rapid identification of SARS-CoV-2 variant after sequencing aids quick response by clinicians to administer relevant drugs and save lives. Also, governments utilize this information to map out strategies for the timely containment of the spread of an identified variant with elevated virulence. The deep learning models reported in this paper show the remarkable predictive results achieved in identifying SARS-CoV-2 strains. However, no work has been done on the identification of recent variants reported globally.
Authors - Ingrid Palma Araujo, Ana Carla Bittencourt Reis, Ari Melo Mariano, and Vinicius Rodrigues Oviedo Abstract - This study proposes to simplify and automate this process, combining two different methods of decision support through multicriteria analysis in a model capable of judging and prioritizing risk criteria in the context of Open Data, presenting the results via iterative online dashboards developed in R. The methodology followed combines the AHP and TOPSIS-2N methods, creating a ranking of the open governmental dataset of the electricity sector in light of the risk criteria evaluated by the proposed model. The AHP technique was used to specify and normalize the importance of each criterion, considering the consistency aspects of the decision matrix. The next step was to apply the TOPSIS-2N method to sort and prioritize these datasets. The results present the datasets that should be improved concerning the respective metadata and the prioritized themes to make the decision-making for the management of the respective bases more agile and assertive.
Authors - Kamal Upreti, Sushma Kumari, Rajesh Kumar, Manmohan Chaudhry, Sandeep Singh, Manpreet Bajwa, Prashant Vats Abstract - Forecasting delivery schedule has always been a basis of urban logistics, but fine-tuning accuracy is now vital for successful outcomes. People's everyday demands have indeed been satisfied by internet meal online delivery services across the globe; for example, platform-to-consumer and steakhouse deliveries in India hit an all-time high of 290 billion orders in 2021 [1]. From of the time a client places an order until it arrives at their door, restaurants must provide correct info about when their meal will be distributed. Offering an estimated time that is longer than the real delivery date would discourage consumers from purchasing, whilst giving a rough guesstimate that is less than the real delivery will boost the number of people who contact our customer service. The major purpose of this study is to propose and provide a OFDA (Online Food Delivery Assistant) establish essential factors for forecasting food delivery batch sizes, as well as to provide a framework for making reliable forecasts. The key impacts and problems of delivery operations in India's various industries are examined and contrasted.
Authors - Marion Adebiyi, Miracle Enwere, Abednego Shekari, Francis Osang, Ayodele Adebiyi Abstract - Deep learning (DL) is an actively growing domain of machine learning owing to the proliferation of big data in multifaceted fields of study. Widespread in literature, there are multiple reports on the implementation of long short-term memory network (LSTM)-based models in the field of bioinformatics for the analysis of the vast amount of genomic data. Genomic data is quintessentially a contiguous stretch of unspaced characters that correspond to the nucleotides adenine (A), guanine (G), cytosine (C), and thymine (T) known as deoxyribonucleic acid (DNA). The DNA sequence must be converted to formats readable by LSTM algorithms run on computers, a process known as digitization. This paper aims to uncover and elucidate the various techniques for digitization in studies that apply LSTM for predictive and classification problems as it relates genomics. We hope this paper would serve as a template for assessing preprocessing techniques for suitability in future studies involving deep learning.
Authors - Yaroslav Yurchyshyn, Vasyl Yurchyshyn Abstract - Inquisitive scientists have been modelling living matter for years, and we can expect to see some breakthroughs in this field. In our paper, we want to convey that this is of concern to us to a certain extent, and we want to touch on what in our opinion is only a small, but very important part of the problem. This is the modelling of blood flow processes. This is not the results of research or experiments. This is how we perceive this process and we want to share our vision with the scientific community. For us, the heart is no longer a pump of the circulatory system, but a complex multifunctional organ that, in a very rough approximation, acts like an escapement (ratchet) mechanism in a mechanical clock, and is a kind of Helmholtz resonator, creating soliton processes in the circulatory system. We propose that the circulatory system of a living organism is treated as a physical system using existing scientific and technical innovations. We look at the possibility of using scientific and technical innovations of physical systems in related medical systems. The possibility of using a new model of the circulatory system of a living organism is discussed.
Authors - Tumelo Presley Nkgapele, Chunling Tu, Moses Olaifa Abstract - Security is crucial in financial institutions to protect clients from financial losses which can occur due to identity theft and ineffective authorization measures. Biometrics techniques are commonly used in these security systems to prevent thieves from stealing people’s identity and using it for financial gains. However, the risk of identity theft has increased over the years due to the development of technology especially the internet of things (IoT) and social media. To secure financial access, various systems and methods have been developed, such as the radio frequency identification (RFID)-based and the biometrics-based automated teller machine (ATM). The RFID-based devices have a dependency on the RFID cards which can be stolen, lost, misplaced, or hacked. Existing bio-metric-based devices also have some drawbacks, with reliability and accuracy as the most common ones. The purpose of this research is to prevent identity theft by using multimodal biometrics technology for financial authorization and authentication. The biometric features used in this paper are the fingerprint and face due to their high accuracy when combined. A more reliable access granting sys-tem is developed with the low possibility to be spoofed. The proposed method also avoids the need to carry an ATM card, therefore, reducing the cases of it being stolen or misplaced.
Authors - Anudeep Arora, Prashant Vats, Gayatri Chopra, Ranjeeta Kaur, Davinder Singh, Manmohan Chaudhary Abstract - An internet crowdsourcing distribution (OCD) technique for on-demand meals is presented in this study. Civilian motorcyclists can be persuaded to act as crowdsourcing employees carrying meals through shared bikes or electrical motorcycles, thanks to World wide web and 3G/4G/5G technology. We present an interactive stochastic optimization system that includes order gathering, solution creation, and sequence delivery operations. To allocate online food duties and generate massive distribution networks in real-time, a mixed meta - heuristic optimization solution meth-od merging the adaptable big neighbourhood searching and socially stigmatized search techniques is developed. Riders from the crowd are constantly distributed among various food vendors. The suggested technique is evaluated using both modeled comparatively tiny and truly huge on-demand online food scenarios. According to the findings, the proposed crowdsourcing food delivery method outperforms typical urban logistics. In less than 120 minutes, the devised mixed optimization mechanism can provide elevated crowdsourcing distribution networks. The findings show that the proposed OCD strategy can help cities with the on-ordering food.
Authors - M. Azani Hasibuan, Rachmadita Andreswari, Muharman Lubis, Fikli Perdana, Hanif Fakhrurroja Abstract - E-learning is one of the technologies used to support the learning pro-cess in University level. Telkom University with the university base on technology, Telkom University so intensively apply the learning using e-learning in or-der to become effective and efficient method. E-learning is used to reduce the cost and time spent learning in an offline class. With E-learning a review system can be done anywhere, because students do not need to come to the classroom to study. Simply by passing the internet network can already attend the class. With the existing data in the e-learning system can be obtained some information by processing the existing data. In this research will be predictions of student data contained in e-learning to determine the performance of algorithms C4.5, Naïve Bayes, and Bayesian Network, which will be used to determine what factors affect the level of graduation of students to a course. In the process will go through several stages of data retrieval, pre-processing, and classification, the process will be done gradually to get the results of all the process.
Authors - Samson Obafemi, Kayode Oyetade, Morvyn Nyakudya Abstract - Daily, latest versions of Information and Communication Technology (ICT) devices are released into the market rendering previous versions obsolete. This calls for the application of green computing to effectively manage the negative impact of these devices on the environment. This paper aims to determine the factors that influence the awareness and practice of ICTs on adoption of green computing in higher education institutions (HEI). The aim was achieved by evaluating academics and students’ level of awareness of green computing as well as determine green computing practices among academics and students. The study adopted quantitative research method. A sample of 125 academics and students from the department of Information Technology, Durban University of Technology were surveyed to test the model developed by this study empirically. The findings found that awareness and practices of green computing has no influence on adoption of green computing on one hand. On the other hand, there is a significant relationship between the knowledge of green computing and the use of computing devices at home, at school, and both at home and in school. Awareness on the safe use and acceptance of environmentally friendly computing practice among staff and students of HEI is recommended by seeking ways it can be integrated into the curriculum. The findings add significant value to our understanding of awareness, practices, and adoption of green computing in HEIs.
Authors - Yuliia Trach, Maryna Tolmach, Olena Chaikovska, Svitlana Khrushch, Kateryna Kotsiubivska, Valeriy Kushnarov Abstract - Using the example of hybrid art, the specifics of the synthesis of technologies and art are revealed as one of the signs and consequences of cultural hybridization. It is emphasized that hybridity is an important principle of modern art, embodying its synthesis with the achievements of science and technology. The specifics of understanding hybrid art in the context of algorithmic and fractal aesthetics are revealed. Thanks to algorithmic aesthetics, the creative process and works of art are considered using computer technologies, methods of applied mathematics and cybernetics. Fractal aesthetics studies fractal art, in which infinity is the main criterion for the generation of a fractal image and the semiotic "underlay" of a fractal picture in general. Bioart is characterized as the result of artists' interest in biotechnology. It is summarized that hybrid art emphasizes technological artistic forecasting, technologies and their impact on man and his world, the future that the artist-scientist reflects on in his works.
Authors - Robi Gusnawan, Muharman Lubis Abstract - In this era of globalization, developments in an area have begun to increase over time both in terms of technology and development, this is indicated by the large number of large-scale projects built by the government and private sectors such as road and bridge construction. Based on data from the Ministry of PUPR in the 2018 Statistical Information Book regarding the allocation of APBN funds to the Ministry of PUPR, shows that the largest APBN allocation is allocated to Directorate General of Highways at 40.04%. The data shows that the highest allocation of funds is allocated to build village road and bridge infrastructure, this indicates that the government is taking priority in carrying out Village Infrastructure development, therefore a project management is needed. In the process of implementing project management, there is often a problem both at the time of planning and when the project is running, things that are commonly experienced by the developer include financial problems such as budgeting, estimation, resources, costs that are not in accordance with the plan, and the embezzlement of funds by unscrupulous persons. These problems can occur due to several important factors such as communication and transparency, hence the project management information system is developed, which aims to facilitate stakeholders in monitoring and ensuring that funds budgeted by the government can be channelled and absorbed optimally, and reports absorption of funds are carried out transparently, to minimize fraud and errors in the village infrastructure development process, by using the Rational Unified Process (RUP) method. The system is designed in the cloud thus all the parties involved in the project can control the entire process and absorption of funds that occur directly so that the infrastructure development process can be right on the target according to the predetermined plan.
Authors - Ingrid Hrga, Marina Ivasic-Kos Abstract - Deep neural network-based systems are increasingly performing tasks that can significantly impact various aspects of human life, such as medical diagnosis or financial risk assessment. Therefore, it is important to understand their decision-making process in order to determine whether those decisions may have been biased. Among the many factors that can influence a model’s decision is that of a data augmentation strategy. In this paper, we analyze how various augmentation techniques affect the image area that the model takes into account when classifying images. We use Integrated Gradients as a method to calculate the attribution of model’s output to its input features. Integrated Gradients results are clustered to determine the strategies the model uses to decide on the label, and to uncover possible changes in decision-making strategy due to the application of a particular augmentation. The results show that even when the accuracy is fairly uniform among the models, there may be a significant difference between the areas of the image to which they give more importance, as a consequence of the applied augmentation.
Authors - David Jennings, Pratheep Kumar Paranthaman, Nikesh Bajaj Abstract - This paper explores the important factors of job interviews that can be gamified and incorporated into a virtual reality interview training application. Virtual reality and gamification have demonstrated benefits in training, simulation, and education. Specifically, virtual reality can be helpful in creating simulations of realistic job interview scenarios and environments for providing interview training for users. Existing interview training applications lack elements of gamification. In this paper, we have proposed a conceptual design of gamified interview training application in virtual reality. In the conceptual design, we have established six stages, each highlighting crucial aspects of interview preparation and training. Currently, a prototype for each stage exists, with each one connected and a user interface system set up.
Authors - Krishna Vaddepalli, Victoria Palacin, Jari Porras, Ari Happonen Abstract - Data quality is key in the success of a citizen science project. Valid datasets serve as evidence for scientific research. Numerous projects have high-lighted the ways in which participatory data collection can cause data quality issues due to human day-to-day practices and biases. Also, these projects have used and reported a myriad of techniques to improve data quality in different contexts. Yet, there is a lack of systematic analyses of these experiences to guide the design and of digital citizen science projects. We mapped 35 data quality issues of 16 digital citizen science projects and proposed a taxonomy with 64 mechanisms to address data quality issues before, during and after the data col-lection in digital citizen science projects. This taxonomy is built upon the analysis of literature reports (N=144), two urban experiments (participants=280), and ex-pert interviews (N=11). Thus, we contribute to advance the development of systematic methods to improve the data quality in digital citizen science projects.
Authors - Lucio Garcia Choque, Luis Jando Galindos Izaguirre, Joel Arturo Flores Canterac, Jose Antonio Ogosi Auqui, Jorge Cano Chuqui Abstract - The globalization of the economy has required companies to implement technology platforms that support the new way of doing business. The use of the Internet for this purpose leads to the development of computer security projects that guarantee the integrity, availability, and accessibility of information. The creation of security policies is a fundamental task that involves the company's people, processes, and resources. Information security allows organizations to protect their financial resources, information systems, reputation, legal status, and other tangible and intangible assets. The main objective of this work is to develop a simulation model that allows evaluating the optimal level of security that society and organizations should have, considering aspects related to risk reduction and obtaining business benefits.
Authors - Jonathan Wilondja Kibulungu, Opeyeolu Timothy Laseinde Abstract - The recent trend of Automation, Industry 4.0 (I4.0), has enabled industries to become more efficient, flexible, smart, autonomous, reliable, and customizable. This new trend improves and replaces existing semi-automatic processes with fully self-controlled systems through its technologies such as smart sensors, PLC (Programmable Logic Controller) and SCADA (Supervision Control And Data Acquisition). Industrial systems or equipment, like a small-scale steam plant, having manual or semi-automatic processes, do not operate effectively as expected due to their actual methods of operation and lack of innovative technologies. Such systems need more attention and monitoring to avoid breakdowns. This research paper aims to demonstrate the reliability and efficiency of a PLC and SCADA based system using I4.0 technologies through a mini steam plant. This research is essential to support the development of I4.0 systems and provide necessary information about the recent trend of automation. The proposed system has advanced features such as sensors, PLC, SCADA, and private cloud. The research results obtained from the LOGO Soft Comfort V6.1 and Wonderware Indusoft Web Studio 8.1, respectively a PLC software and SCADA software, confirm that the mini steam can operate automatically and efficiently and can be well monitored.
Professor, Nirma University, Institute of Management, India
Dr Nityesh Bhatt is a Professor & Chair (Information Management Area). He has more than 22 years of experience in academia, corporate training and research. In 1998, he was awarded as the best faculty of NIIT in North India. Credited with good number of research papers & management... Read More →
Saturday August 27, 2022 1:16pm - 1:17pm BST
Virtual Room BLondon, United Kingdom
Authors - Vivek Banerjee, Bhagwat kakde Abstract - The growing demands of communication network bandwidth for emerging technology raised the issue of spectrum scarcity. The management and allocation of spectrum in wireless communication are major challenges. A cognitive radio network can help alleviate a wireless network's spectrum shortage. The optimal allocation of resources in cognitive radio networks boosts the wireless communication system's capability. Machine learning was employed to enable optimal resource allocation in cognitive radio networks in this study. The proposed power allocation strategy for a cognitive radio network. Because of its low processing overhead, the suggested technique is referred to as lightweight resource allocation in the cognitive radio network. The MLP network is coupled with a cognitive radio network channel in the suggested machine learning technique. In a cognitive radio network, the proposed technique increases spectrum use. MATLAB tools are used to evaluate the proposed technique against RLA and CMA resource optimization strategies. The results reveal that the suggested strategy for distributing resources is extremely effective.
Authors - Luisa Paola Jorqueda Vasquez, Luciana Victoria Jurado, Jesus Lagos Moron, Mario Reynaldo Medina Ramos, Jose Antonio Ogosi Auqui Abstract - Academic management is based on being able to facilitate and improve the different processes in the institutions, in order to have a successful pro-cess, but many of these do not have an adequate documentary procedure that accompanies it at the time of carrying out the present processes, that is why with theseantecedents a computer system was raised and developed that allows us to improve the management process in the documentary process to At the time of this being required, the system that was developed helped to improve this process thus reducing the established response time of each query that was had with the client. For the development of the computer system, the RUP methodology was used, in addition to the use of MySQL for data storage and Visual Basic for the realization of the system. On this it was possible to observe the decrease in time in the response that was had with the client when looking for any type of procedure.
Authors - Fahdi Saidi Lubis, Muharman Lubis, Lukmanul Hakim, Hanif Fakhrurroja Abstract - Currently, the establishment of legal act in cyberspace is the inevitable consequence of globalization to ensure the legal certainty of the people who interacting within internet. With the number of legal infringement in cyberspace increasing each day, the needs of Internet regulation has becoming a world concern. Indonesia is one of top ranking country with high number of internet users, which want to safeguard the public interest with a comprehensive law through “Undang-Undang Informasi dan Transaksi Electronik” (UU ITE). Therefore, the current use of this act is retrieving the diverse feedback and responses from citizens in Indonesia since it was signed or after the amendment for several aspects. The availability of social media networking enables them to react to the subject concerning legal issues within internet. Thus, this research work used text mining technique to find out the perceived issues, challenges, and benefits of ITE implementation from citizens’ opinions, comments, and perceptions in social media of Twitter. The result of this study grouped into two main topics; legal consequences and precedence.
Authors - Maryori Dayana Flores Ventes, Alex Leonardo Quispe Matta, Edwin Abel Cabana Palomino, Jose Antonio Ogosi Auqui, Jorge Cano Chuqui Abstract - Augmented Reality is a technology that has appeared in recent years, having a significant impact on the lives of many people.
Authors - Saptorshi Bhattacharjee, Kishore Kumar Senapati Abstract - In the 21st century as the technology evolves in a rapid pace, more and more number of users and businesses are getting connected to the internet. A vast majority of these users use the wireless mode (Wi-Fi) for getting connected to the internet. As the number of users and the online transactions rises in billions so is the risk of a cyber-fraud, due to the vulnerabilities present in the Wi-Fi protocols (WEP, WPA and WPA2). Security remains a huge challenge as Wi-Fi passwords could be cracked from both home and enterprise level networks by using the tools and by performing social engineering attacks. Data protection is a huge challenge for computer scientists and in this work the investigators have identified the loopholes of Wi-Fi encryption protocols. The identification procedures are tested on a test network with various Wi-Fi encryption protocols in a controlled laboratory setup. Data packets from the wireless networks could be easily sniffed by using aircracking software suite for capturing the handshake and ultimately for cracking the password from huge wordlists. The process of cracking the password could also be speed up by using a Graphics Processing Unit (GPU) which performs set of repetitive tasks in a faster way and helps in extracting passwords. The overall investigation helps investigators to understand the vulnerabilities present in the modern-day Wi-Fi protocols. Hence it has been proposed the use of RADIUS server model like in an enterprise network and the use of Wi-Max standard of security for better safety measures.
Authors - Anand Gachhadar, Ram Krishna Maharjan, Surendra Shrestha, Nanda Bikram Adhikari Abstract - In this paper we propose a mathematical framework of throughput for multi-tier heterogeneous network involving cell load factors, noise figure and data rate which measures the resource consumption. Due to increase in wireless networks, interference among the networks has also been increased which motivates the use of interference mitigation technique. Hence a statistical framework of throughput analysis in 5G has been presented to mitigate the interference using successive interference cancellation (SIC) technique in heterogeneous networks and boost the throughput gain. Interference at various tiers degrades the throughput of the system. Throughput gain for both cellular and D2D user is calculated and effect of interference cancellation technique is analysed. The results show that network throughput gain can be significantly increased by the use of better interference cancellation technique and spatially separated antennas. Throughput gain is measured w.r.t user density, load factor, data rate and noise figure at fixed SIRT for with and without using cancellation techniques.
Authors - Ilham IHOUME, Rachid TADILI, Nora ARBAOUI, Mohamed BENCHRIFA, Ahmed IDRISSI, Mohamed DAOUDI Abstract - Equipping the agricultural industry with state-of-the-art technologies such as big data analytics, cloud computing, and the internet of things (IoT) is one of the most promising solutions to more efficient and sustainable yield production. In this context, we develop a tinyML-oriented model for a deep learning-based greenhouse microclimate management to be integrated into an onfield microcontroller. Multivariate climate data were collected from sensors installed inside a designed experimental strawberry greenhouse. The obtained values’ combinations were labeled according to an eight-action control strategy, then used to prepare a balanced deep learning-ready dataset. The latter was used to train, cross-validate, and test 90 Multi-Layer Perceptrons (MLPs) with varied hyper parameters to select the most performant and optimized model instance for the addressed task. The final selected model incorporates one hidden layer with 8 neurons and has just 104 parameters; it scored a mean accuracy of 95% during the cross-validation phase and 94% on our supplementary test set. The model enables active, robust, and autonomous greenhouse management with the less required computations. It can be efficiently deployed in microcontrollers within real-world operating conditions.
Authors - Dmitry O. Batrakov, Mariya S. Antyufeyeva, Angelika G. Batrakova Abstract - This paper presents the results of extended approaches to processing signals of pulse ultra-wideband (UWB) ground-penetrating radar (GPR). These approaches are based on improving the algorithms of inverse scattering problem solutions. From applying point of view, the case, when the studying construction consists of a finite number of layers whose thickness is commensurate with the spatial size of the probing pulse, is very important. For this instance, the procedure of numerical correction of the reference (probing) signal is proposed. Using the results of this rectification for creating the algorithms of inverse problem solutions allowed to propose the new technique of thickness measurement problem. The obtained data are based, on the one hand, on the results of real laboratory experiments, and, on the other hand, on the results of numerical simulation. The presented graphs demonstrate the effectiveness of the proposed approach. The reached results are important and relevant primarily for technical applications related to the non-destructive testing of the current condition of critical infrastructure objects, primarily roads and bridges.
Authors - Avon Budiyono, Zenitho Madyagantang Hakiki, Ahmad Almaarif, Muharman Lubis, Muhammad Bambang Hidayanto Abstract - Data centre plays a big role for the existence infrastructure of information technology to handle data processing, providing telecommunication and network services. The performance relies on the operational temperature and humidity from the computers and its related equipment in the server room that needs to be maintained by the status from the air conditioner used in the server room, which must be always noticed by the operator. Consultation Company is an in-formation and technology company that provides a service in data centre. This company, however, has not yet implemented a cooling system sensor in the server room to measure the temperature and humidity. The stated condition for the problem had led the design and development of a prototype cooling system sensor in measuring the temperature and humidity values for the approach of monitoring process as the solution. PPDIOO Life-Cycle Approach is used as a research method within the step of designing and developing this prototype. The sensory device for this prototype is using DHT11 sensor that is attached in microcontroller Arduino along with ESP8266 for the wireless connection module. The information has been monitored from an application called Blynk that can inform the value of the temperature and humidity.
Authors - Oktay ERISIK, Ekrem Faruk OZKAN1, Erman OZDERLI, Turan TURHAN, and Dogancan GUNAL, Hasan Basri CETINKAYA, Bayram KURUCAY, Recep TETIK Abstract - Industrial facilities supply their electrical energy demand widely through the national electricity grid. On the other hand, critical industrial facilities have their own energy generation resources besides the grid connection for energy reliability. Gas turbines and steam turbines are generally preferred as electricity generation sources for critical industrial plants. Recently, studies have been carried out to use wind power plants and energy storage systems to support these conventional generation units, especially in order to reduce the carbon foot-print of industrial facility. Critical industrial facilities having their own generation mostly prefer an operating structure parallel to the electricity grid in order to ensure the continuity of their energy supply in case of severe faults in their internal systems. However, it is difficult to cope with a huge grid facing with severe faults. Since the continuity of energy supply is critical, certain grid failures must be isolated quickly and safely. Decoupling system are installed in order to meet this isolation from the grid during severe faults. In this study, the decoupling system model is presented for a refinery with a built-in power plant and also connected to a 154kV transmission system. The philosophy of decoupling system, as well as hardwire and software structure have been discussed in detail. Information exchange with other systems that are important for the reliable operation of the facility have already been investigated.
Authors - Osama Aledamat, Noureldin Elshiekh, Ahmed ALSalahi, Loay Ismail Abstract - Many countries worldwide are taking steps to reduce air pollution caused by internal combustion engine (ICE) vehicles; leading the popularity of electric vehicles (EVs) to bloom. As the number of EVs keeps increasing, more demands are expected to be placed on the charging station infrastructure to adapt to this increase. In this paper, a system for monitoring and managing EV charging is proposed. For the system, three modules are presented, the EV module, the charging station module, and the charging point module. Each module is shown with both its hardware and software designs and its internal systems. The hard-ware implementations for the three modules are presented with the software implementation for the mobile application and the state-of-charge (SoC) estimation model. Finally, the system is tested for both the SoC estimation and the mobile application.
Authors - Ali S. Kamal, Syed M. A. H. Bukhari, Muhammad U. S. Khan, Tahir Maqsood, Muhammad A B Fayyaz Abstract - Most of the internet traffic is encrypted and it is a challenge to identify streaming videos in the network traffic. The present paper will present a methodology named as Traffic Pattern Plot (TPP) to identify video streams in encrypted network traffic. The proposed methodology plots the traffic flows of videos and uses a Convolutional Neural Network to identify the videos. The results show that the traffic pattern plot generated from 120 seconds of sniffing network traffic is enough to identify the video even in the encrypted network traffic with 94% accuracy.
Meelis Kitsing is the Rector of and Professor of Political Economy at the Estonian Business School. Previously, Kitsing worked as Head of Research at Foresight Center, a think-tank at the Estonian Parliament, an Adviser at the Strategy Unit of the Estonian Government Office and Head... Read More →
Saturday August 27, 2022 3:31pm - 3:32pm BST
Virtual Room ALondon, United Kingdom
Authors - Petya Bicheva, Evgeni Valchev Abstract - The dynamics which the business environment and business processes develop in modern conditions require minimizing threats from both external and internal factors. In this article we present a different point of view - the ability to overcome major problems through the usage of intelligent risk analysis. It com-bines the established stages of the risk management process according to ISO 31000 (Risk Management) and the capabilities of artificial intelligence. The contribution of this paper is in the mechanism of acquiring added value through the usage of artificial intelligence. This’s the way, the risk analysis of business pro-cesses can process extremely large amounts of data in real time; to automate complex and time-consuming risk management processes; to respond optimally and adequately to current threats. The purpose of this paper is to offer a new look at risk analysis methodology. Examining the process and going through all stages of management, reveals the possibilities of the concept / technology to predict and react INTELLIGENTLY to respond long before their occurrence; to help turn threats into opportunities. In order to prove the applicability of the method-ology, we consider various cases and follow their development, with and without the participation of artificial intelligence. All this is aimed at reducing the frequency and severity of risk events in the management of business processes.
Authors - William Elvis Alarcon Cotrina, Modesto Emerson Delgado Canales, Enrique Michael Gutierrez Marino, Diego Andres Crespo Buquich,Jose Antonio Ogosi Auqui Abstract - With the development of this article we want to present the information in a more detailed and concise way on the subject of computer security and the auditorias in such a way, it is easy to enterintoany public that stumbles upon our article. For its development, a compilation of information was made from different articles, books and research on the subject of computer security and auditing. A proposal was made of what are the phases or the essential procedure to carry out an audit in a correct way that will help the people who are starting with the audits.
Authors - Yahya ZEFRI, Imane SEBARI, Hicham HAJJI, Ghassane ANIBA Abstract - The extraction of photovoltaic (PV) arrays from remotely sensed aerial imagery is the core element in the mapping of PV sites. Moreover, it is of high utility in increasing the robustness of automatic imagery-based inspection models through reducing their searching space. The extraction can also be used reliably as a tool in assessing the generation capacity of PV installations. In this context, we leverage the channel-attention deep learning mechanism to exploit the full potential of the RGB spectrum and develop a deep semantic segmentation model for the extraction of PV arrays from very high spatial resolution imagery acquired by Unmanned Aerial Vehicles (UAVs). We collect aerial images from 4 large-scale PV sites featuring multi-type PV arrays in various arrangements and backgrounds. After a photogrammetric post-acquisition workflow, we prepare a labeled big imagery dataset comprising 7319 unique 256×256 px tiles with a spatial resolution around 1 cm/px. Our final model relies on a hybrid LinkNet architecture embedding a Squeeze-and-Excitation (SE) 18-layer residual network backbone. It operates on monocrystalline, polycrystalline and thin-film surfaces, on which it achieves the state-of-the-art performance, with a mean IoU of 86.67% and F1-score of 92.20% on our test set.
Authors - Hassam Elzayady, Mohamed S. Mohamed, Khaled Badran, Gouda Salama, Ahmed Abdel-Rahim Abstract - The spread of hate speech has become particularly apparent on social media platforms as their usage for communication and unrestricted thought has increased on a worldwide scale. However, this could cause disagreement and antagonism among users, creating an unattractive online environment. Countries, companies, and academic institutions have all invested heavily in finding an effective solution to this challenge. There has been less study done in Arabic compared to other languages to develop automated systems for recognizing hate speech. Additionally, Arabic research on the correlation between personality traits and hate speech still remains rather limited. In this paper, we propose a novel method for enriching MARBERT model that incorporates static word embedding (AraVec 2.0) and personality trait features for Arabic hate speech detection. The experimental results indicate that the suggested methodology exceeds in terms of macro-F1 score by achieving 86.4 % compared to previous re-search reported in the literature.
Authors - Alaa M. Odeh, Isam Ishaq Abstract - One of the foremost concerns for each family is to keep their home secure and safe for living. And perhaps this is what we keep thinking about all day. For this reason, this paper introduces a designed Internet of Things (IoT) home security system. The proposed work contains three main circuits. Each contains a sensor and an actuator that work based on the data received by the sensor. These three circuits are: a temperature monitoring circuit to monitor the temperature of a room and activate the fan when needed, a gas and smoke detection circuit to monitor gas leakage and notify when gas or smoke is detected, and a touch detection circuit to secure home from robbery and unauthorized entry. All these circuits are connected to an ESP32 controller which controls their working and sends continuous sensors' readings to the ThingSpeak cloud. The paper presents a complete hardware and software description of the system. In addition to some followed security settings to ensure secure data transmission and reception.
Authors - Pratheep Kumar Paranthaman, Srikant Vallabhajosula, Alys Giordano, Stacey Walton, Oliver Tuisa, Nikesh Bajaj Abstract - Lower body tracking and user experience analysis are ongoing challenges in immersive technologies and biomechanics research, respectively. Particularly, in immersive technologies like mixed reality there are limited research applications in user comfort and experience, and locomotion. In this paper, we aimed to conduct a cross-disciplinary research (by combining the elements of physical therapy and human computer interaction) in understanding the impact of user experience and immersive technologies induced symptoms in virtual and mixed reality environments for obstacle clearance tasks. For this, we created obstacle clearance applications for Oculus Quest and Microsoft HoloLens. Also, we conducted a user study with 12 participants under two conditions (virtual reality and mixed reality) in obstacle clearance behaviour of varying heights. The results from our study indicated that the levels of immersion and enjoyment were significantly higher in virtual reality compared to mixed reality. In terms of virtual reality induced symptoms, the participants found the mixed reality environment to be slightly more comfortable than virtual reality.
Authors - Matheus Alcantara de Santana, Claudio de Souza Baptista, Andre Luiz Firmino Alves, Anderson Almeida Firmino, Gerson da Silva Januario, Roney Wellington da Silva Caldera Abstract - Driven by technological advances that make the purchase process more comfortable and efficient, e-commerce has grown every year, thus increasing the number of internet sales. Because of the large volume of products available from different suppliers, it becomes difficult to identify and group the same products to find the best deals. One of the most significant barriers is the existence of various descriptions of the same products. In this work, we use product matching with a heuristic created to help identify similar products to solve this problem. We employ several machine learning models, including BERT, and use a database of invoices from Brazil to carry out the experiments. At the end of the implementation, the model could classify the invalid products. The results of the classification task showed that our proposed solution is promising, with an f1 score of 92%.
Authors - Jagdeep Singh, Shivoham Singh, Mamta Kumari, Surendra Kumar Vyas Abstract - In the era of covid-19, most of the business declined & a huge loss of jobs due to no demand and automobile sector is not the exception. The auto sector contributes close to 7.5 percent of the total GDP in India where SCM is one of the key contributors to overall value creation for an organization. The efficient & effective supply chain is dependent on information & communication technologies (ICTs) at present and hence it implies that ICT is spine of SCM. The aim of the research was to draw the conclusion on impacts of ICT practices on SCM performance in automobile industry in India. The outcomes state that ICT practices has high correlation & direct impact on SCM performance however it does not have much impact on operational performance. The research also suggested that better & more effective ICT practices result better supply chain performance. The limitation of the research was the unpaid cooperation of the respondents and the ICT practices are limited to supply chain operational performance in various departments & functions only & not the applications in any vehicles.
Authors - Victor Hugo Guadalupe Mori, Jose Antonio Ogosi Auqui, David Hugo Obando Pacheco, Jorge Cano Chuqui Abstract - Currently, companies have the obligation to acquire new technologies to automate their processes, activities and functions, so that they can guarantee an efficient service. In this article we present a proposal to implement a web system to improve the digital marketing of the budamaki company thus fulfilling one of its objectives such as developing high quality products and acceptance for its customers. So we opted for the development of a web page for greater inter-action with your customers.
Authors - Assia Najm, Abdelali Zakrani, Abdelaziz Marzak Abstract - There is evidence that precise prediction of the software development effort plays a crucial role in properly monitoring and managing software projects. Researchers have suggested numerous techniques for accurately estimating software effort. However, to date, no large-scale studies have been undertaken to examine the prevalence of interpretability of estimates. In particular, in the literature, the support vector regression model has shown good performance in estimating both agile and non-agile software development costs. Despite its remarkable outperformance, this black-box model needs to be interpreted using several model-agnostic explanation techniques. This study aims to explain the convenience of model-agnostic methods in interpreting the estimated effort using the SVR-RBF model optimized with an Artificial Immune Network in an agile and non-agile context. The estimations are performed using two datasets, the leave-one-out cross-validation and 30% holdout methods. To help practitioners and managers trust and comprehend the SVR estimates, we employ four global (Feature Importance, Interaction, Dependence, and Summary Plot) and two local interpretability methods (Force and Waterfall Plot). The results show that global and local interpretability techniques effectively assist practitioners in decision-making.
Authors - Victor Hugo Guadalupe Mori, Jose Antonio Ogosi Auqui, Jimmy Aurelio Rosales Huamani, Jose Luis Arenas Niquin Abstract - The term Tic's, in this context, refers to the information systems that infer in the data process, allowing its easy access and distribution, thanks to automation through databases and storing in trusted servers, this also refers to the media such as the use of digital marketing through social networks, advertising through advertisements on television or radio stations, for which it is widely used by companies that want to grow, allowing them to have a better image and better quality of their services In a business context, information systems generally serve the purposes of any other system, such as: data processing, entity data storage, reporting, and other types of data summary tools. However, in the era of digital transformation, the use of information systems, especially those related to management, plays an important role in ensuring the technological integration of all operations. Service management gives the company access to the right knowledge to make quick and accurate decisions.
Authors - Pelumi O. Bamgboye, Marion O. Adebiyi, Abayomi A. Adebiyi, Francis B. Osang, Ayodele A. Adebiyi, Miracle Enwere, Abednego Shekari Abstract - Customer review is constantly generated in the form of text based on customer opinion concerning specific products in the electronic (e)-commerce space. With the proliferation of data, the classification of customers’ opinions is a major issue in e-commerce. However, merchants frequently require customers to input their opinion about a product, in order to harvest customer experiences. The text classification technique helps customers and merchants to know if a product has a positive or negative review. This will enable merchants to improve the customer experience and also improve the revenue of the company. Sentiment analysis was executed on a customer product review dataset using support vector machine (SVM) in this study. The experimental results obtained show an accuracy of 86.67% in classifying customers’ opinions on selected products. The findings will guide merchants to know customer feedback and identify their needs to improve sales culminating in increased revenue.
