		FeedBack \| Contact Us \| Links \| Site Map

Welcome to Journal of Theoretical and Applied Information Technology

Submit Paper / Call for Papers

Journal receives papers in continuous flow and we will consider articles from a wide range of Information Technology disciplines encompassing the most basic research to the most innovative technologies. Please submit your papers electronically to our submission system at http://jatit.org/submit_paper.php in an MSWord, Pdf or compatible format so that they may be evaluated for publication in the upcoming issue. This journal uses a blinded review process; please remember to include all your personal identifiable information in the manuscript before submitting it for review, we will edit the necessary information at our side. Submissions to JATIT should be full research / review papers (properly indicated below main title).

Journal of Theoretical and Applied Information Technology

August 2023 | Vol. 101 No.15

Title:	EVALUATION OF ERGONOMICS OF AN INDUSTRIAL ZONE BASED ON FUZZY COMPREHENSIVE EVALUATION METHOD
Author:	EL M’HADI HAJAR , CHERKAOUI ABDELGHANI
Abstract:	In this paper, we will evaluate the level of ergonomics in an existing industrial zone. This evaluation will enable the community to ensure that the industrial zone meets the expectations of the companies and the involved actors, to identify improvements track and to initiate actions. The aim of this approach is to improve the quality of the area, to give it the maximum chance of perpetuating direct and indirect jobs and to preserve environmental resources (water, energy, landscape, air, etc.).To meet this evaluation objective, we proposed a model based on the combination of the AHP method and the fuzzy comprehensive technical evaluation. Considering the fuzzy characteristics of factors influencing the level of urban ergonomics in an industrial zone, a two-level evaluation index system was established. The degree of adherence of factors is determined by the knowledge and experience of the experts. Then the weight of each factor is determined by the AHP method and a final judgment matrix is determined by the end. This method allows us a quantitative ergonomic evaluation.
Keywords:	Fuzzy sets, AHP Theory, Industrial Zone, Ergonomics Evaluation.
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	FINANCIAL TECHNOLOGY ADOPTION: A CROSS COUNTRY STUDY
Author:	MARSHEILLA ANAK VICTOR, MOCHAMMAD DODDY ARIEFIANTO, RINDANG WIDURI
Abstract:	This study aims to examine how the relationship and influence of variables named Facilitating Conditions of Financial Sector, Facilitating Conditions of non Financial Sector, Overall Valuation on Business Climate, and Social Conditions and GDP as a control variable towards the adoption of financial technology across countries using some modifications of the UTAUT framework as novelty. The data were sourced from the Global Financial Inclusion Index, the Global Financial Development Database and the World Development Indicator on the World Bank website in research periods of 2011, 2014, 2017 and 2021 with a total of 500 observations from 125 countries. Panel data econometric techniques employed are pooled least square, fixed effect and random effect model. We find that financial technology adoption is positively associated with the financial sector and overall valuation on business climate. While social conditions have a negative significant effect, the non financial sector has no effect on financial technology adoption. Financial sector intervention through financial technology has become a new instrument which triggers financial growth and helps realize financial inclusion more quickly. Financial inclusion is one element of financial growth and development which is generally measured by the scope and ease of access to financial services in a country.
Keywords:	FinTech, Economic Growth, UTAUT, across Countries, Panel Regression
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	TASK SCHEDULING IN CLOUD COMPUTING: PRIORITY-BASED ALGORITHMS AND FUTURE DIRECTIONS IN DATA SCIENCE
Author:	RASHA AL BASHAIREH
Abstract:	In the era of distributed computing, Cloud Computing (CC) has been considered an emerging technology that delivers on-demand services using the worldwide online medium. One of this technology's main challenges is scheduling tasks and allocating resources to achieve the best performance with minimal optimized execution time, and less resource time and usage. Thus, the importance of task priority has been recognized by many researchers since tasks submitted to the cloud can have various sizes, resource utilization, and execution times. Therefore, existing task scheduling algorithms prioritize tasks, where the task with the highest priority is allocated to the available resource. This paper investigated the task priority problem in cloud-based Task Scheduling (TS) algorithms via three different directions centered around the following contributions; first, a fruitful discussion and comparison of some selected cloud priority-based TS algorithms regarding their scheduling method and parameters, algorithm performance, and limitations reflecting how the priority issue is handled in the cloud environment. In the Data Science and Artificial Intelligence age, the interaction and intervention between the Big Data sector and CC are gradually advancing due to mutual development. Cloud environments have become a prominent platform for big data applications, forcing the integration between Data Science and CC techniques to evolve. On this basis, secondly, the paper addresses the directions of potential future research of cloud priority-based TS schemes in Data Science. Finally, as a result, a conceptual framework incorporating a task content-type component is proposed to support the processing and scheduling of data-related tasks submitted to the cloud.
Keywords:	Cloud Computing, Task Scheduling, Priority Algorithms, Data Science, Artificial Intelligence
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	HYBRID APPROACH FOR MOVIE RECOMMENDATION SYSTEM USING DOUBLE COLLABORATIVE FILTERING AND SEQUENTIAL PATTERN ANALYSIS
Author:	HARRY YANTO, SANI MUHAMAD ISA
Abstract:	Recommendation systems (RecSys) are essential nowadays to handle information overloading. RecSys algorithms usually predict whether the user will like the content or not based on the previous consumed content by the user. User’s preferences will be changing from time to time, but the research of RecSys algorithm nowadays rarely involves the sequence of user preferences. We also trying to engage two Collaborative Filtering algorithms (CF) with different attributes. CF that use the rating as attribute (CFR), and CF that use the user-preferred genres as attribute (CFG). We hybrid those three algorithms CFR, CFG, and Sequential Pattern Analysis (SPA) to increase the accuracy of the recommendation system. Then we evaluate it using the f1 score and compared it to the CFR, CFG, and SPA alone. This research concludes that hybrid CFR, CFG, and SPA increase the accuracy of f1 and precision score compared to the algorithm stands alone. We also conclude that cosine is the best similarity to use in searching similar users for RecSys
Keywords:	Recommendation System, Hybrid, Collaborative Filtering, Sequence Pattern Analysis, Best Similarity
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	FAILSAFE ALGORITHM MANAGEMENT IN SMART FARMING IOT CONTROL SYSTEMS
Author:	MUSTAFA MAN, TOH WEI JER, AEZWANI WAN ABU BAKAR
Abstract:	Internet of Things (IoT) technology is an important component in the modern smart farming system and plays a key role in improving the efficiency of various aspects of the farm such as greenhouse troubleshooting, farm data tracking, smart farm control and many more. Previous research has established that with the help of IoT technologies, who will be in-charge of the data monitoring of the farm, owners are able to perform necessary farming tasks including seeding, fertilising, watering and so on at the perfect moment to ensure maximum agricultural yield. Multiple sensors are used to measure the crucial environment parameters are implemented to collect the farm data and transfers to the backend server by the IoT module in order to perform necessary data analysis thus completing the IoT usage in the given farm. The reliability and time accuracy of the data collected by the sensor is crucial for the following control system. To date, only a limited number of Smart Farming IoT system architectures have been identified, as well as the failsafe data management issues of the IoT modules in designated farm types is yet to be resolved. this paper is that researchers develop a data failsafe management algorithm compatible with Arduino embedded system platform which works together with backend server to manage and transfer the data received from the sensors autonomously. Unlike conventional smart farming IoT architecture, the proposed system is equipped with multiple data storage media located in different parts of the system preventing data losses during unwanted occasions. This review paper talk about implementing failsafe architecture into IoT smart farming systems. Besides, the system proposed in this research could fit with various wireless connectivity arrangements while ensuring the smoothness of the network traffic. Expected results showed that the proposed IoT architecture resolved the data loss issues during the internet outage thus improved the IoT control system. Therefore, all the benefits brought to the farm by implementing IoT technologies are well preserved.
Keywords:	Smart Farming, Internet of Things (IoT), ESP32, IoT Connectivity, IoT Failsafe Algorithm, Failsafe Data Management
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	CREDIT CARD FRAUD DETECTION USING MACHINE LEARNING ALGORITHM
Author:	MOH. ALIANSYAH, FERALIANA AUDIA UTAMI, RIYANTO JAYADI
Abstract:	Credit card fraud is a serious issue in the financial services industry. Billions of dollars are lost each year due to credit card fraud. Credit cards have become one of the fastest growing financial services by banks in recent years. However, with the increasing number of credit card users, banks face an increasing rate of credit card defaults. Credit card fraud is related to the use of prohibited credit card materials for acquisition. The aim of this research is to analyze and compare the usage of three types of machine learning algorithms consisting of Neural Network (NN), K-Nearest Neighbour (KNN), and Random Forest (RF) to analyze and predict credit card fraud security. The credit card transaction dataset was sourced from European cardholders and contained 284,807 transactions. The results of the analysis state that the Neural Network Algorithm has the highest prediction capability (99.80%) and precision rate (100%) compared to the other two machine learning algorithms.
Keywords:	Fraud Detection, Machine Learning, Neural Network, Random Forest, K-Nearest Neighbor
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	INTELLIGENT COLLABORATIVE FILTERING RECOMMENDATION SYSTEM FOR MOVIE REVIEW RATING
Author:	SRINIVASA RAO MANDALAPU, B. NARAYANAN, SUDHAKAR PUTHETI
Abstract:	Recommendation system used to analyze the data concept and discovery of data to provide the modified approvals on the web. Various techniques have been suggested to capture the user's interest and effectively provide accurate recommendations. User-based collaborative filtering technique was popular and utilized in practice extensively. Still, the system faced several vital tasks to provide adequate scalability and qualified recommendations because of increasing items daily and users on various webpage. So, a novel technique named Eagle Deep Neural based Movie Recommender System (EDNbMRS) to recommend movies related to ratings from the review. Consequently, the data was collected for the training process. In the preprocessing phase, the noise features were eliminated from the dataset for the recommendation process. Next, the best features were selected based on the developed model in the feature extraction process. Then, a rating for the movie was necessary for recommending the film to the user to perform the similarity and prediction computation process. Finally, the recommendation process was done using the predicted rating score for the target user. Meanwhile, the developed mechanism was executed in the Python tool with several performance measurements such as accuracy, Recall, f-measure and precision. This developed model produced outstanding results compared with the previous studies by providing accurate results for the recommendation process.
Keywords:	Recommendation, Target User, Feature Analysis, Eagle Optimization, Collaborative Filtering
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	INTEGRATION OF NON-HIERARCHY CLUSTER ANALYSIS WITH SEMIPARAMETRIC TRUNCATED SPLINE MULTIRESPONSE REGRESSION
Author:	EVITARI GALU ERWINDA, SOLIMUN, ADJI ACHMAD RINALDO FERNANDES, RAHMA FITRIANI, ATIEK IRIANY
Abstract:	Cluster analysis is an approach used to find similarities in data and place them in groups. Cluster analysis is divided into two methods, namely hierarchical and non-hierarchical methods. The non-hierarchical method of several methods, namely K-Means, K-Harmonic Means, and K-Medoids, used in this study is K-Means. Then, the integration of cluster analysis with semiparametric truncated spline multiresponse regression will be carried out. This study aimed to identify and develop non-hierarchical cluster integration and semiparametric multiresponse regression using the Truncated Spline approach in the case of Farmer Inclusiveness data in West Nusa Tenggara. Modeling factors affecting the inclusiveness of farmers in NTB was conducted using cluster analysis modeling with multiresponse regression which was developed by making integration of non-hierarchical cluster analysis with semiparametric truncated spline multiresponse regression to help give Comprehensive information about farmer inclusiveness in NTB. The sample used in this study was 100 farmer groups in NTB. The results in this study produce three clusters, which have low, medium, and high inclusiveness categories. Cluster 1 is a farmer with a low inclusiveness category consisting of 42 farmer groups, cluster 2 is a farmer with a medium inclusiveness category consisting of 21 farmer groups, and Cluster 3 is a farmer with a high inclusivity category consisting of 37 farmer groups. The cluster with the highest R-squared on modeling integration of non-hierarchical cluster analysis with semiparametric truncated spline multiresponse regression is cluster 1 with R2 value of 0.205. Thus, modeling using cluster integration with semiparametric truncated spline multiresponse regression shows a change in the nature of the data that has a variable relationship pattern and can help the government and farmers to determine the level of inclusiveness of farmers and examine the factors that influence farmer inclusiveness. This research’s originality is combining cluster analysis with semiparametric truncated spline multiresponse regression with integration between the two methods.
Keywords:	Integration Clusters, Multiresponse Regression Analysis Semiparametric, Truncated Spline, Farmer Inclusivity.
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	A FUZZY INFERENCE MODEL FOR DIAGNOSIS OF DIABETES AND LEVEL OF CARE
Author:	TEH NORANIS MOHD ARIS, AZURALIZA ABU BAKAR, NORMADIAH MAHIDDIN, MASLINA ZOLKEPLI
Abstract:	Diagnosis of diabetes is a complex decision-making process. The creation of diabetes diagnosis models is vital in the decision-making process and requires adequate information for fast detection and treatment. Diabetes is detected from a set of symptoms. The symptoms data are an important reference to diagnose diabetes which are collected and stored in datasets. Diabetes datasets are prone to vagueness and uncertainty. In addition, insufficient information on the diagnosis of diabetes exists and this problem is not addressed in previous research. This research work analyzes a simulated diabetes treatments dataset that were validated by medical expert [1]. A new fuzzy inference model based on Mamdani method is designed to provide interpretable understanding and sufficient information on diabetes diagnosis which is combied with the level of care to support the vagueness, uncertainty, and insufficient information problems.
Keywords:	Decision-Making, Fuzzy Inference Model, Dataset, Diagnosis Of Diabetes, Level Of Care
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	RURAL UNIVERSITY STUDENTS’ CONTINUOUS USE OF ONLINE PROCTORED EXAMINATIONS
Author:	ADMIRE CHIBISA, DAVID MUTAMBARA
Abstract:	The outbreak of the Covid-19 pandemic caused untold suffering and calamities. Many nations introduced national lockdowns as a way of curbing the spread of the virus. All services were halted except essential services. Unfortunately, higher education and training was considered non-essential and hence stopped. However, the academic project had to continue such that academic institutions had to find other means of operating. Online learning and online assessments were introduced and used during the pandemic. The purpose of this study was to explain rural university students’ continuous use of online proctored examinations in the aftermath of the pandemic. The study used the Expectation Confirmation Model to develop the “Continuous Use of Online Proctoring Software Model” where the online proctored software was used to shadow the use of online proctored examinations. A cross-sectional survey was used to gather data from a sample of 335 respondents of which analysis was done using Partial Least Squares Structural Equation Modeling, with the help of SmartPLS version 3. The developed model explained 69,2% of the variance in continuous use of online proctored examinations. These results showed that rural university students were in favor of continued use of proctored examinations. The study recommended that university authorities should consider the continued use of online proctored examinations though this comes at an increased cost of students’ access to relevant devices and connectivity.
Keywords:	Continuous Use, Online Proctored Examinations, Online Proctoring Software, Rural University Students, Expectation Confirmation Model
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	QLAR- Q LEARNING BASED ACOUSTIC ROUTING FOR UNDERWATER SENSOR NETWORKS
Author:	PRATHIBA N, MALA C S
Abstract:	Underwater sensor networks have gained huge attention due to their significant use in monitoring and exploring the ocean, lakes, seas and rivers. The UWSNs are different from the ground based static sensor networks. The offshore networks or the networks which are deployed on portion of land have been explored widely but implementing the traditional protocols is not considered as a feasible solution because these networks suffer from several challenges such as high- water pressure, low bandwidth, delay and error rate etc. Therefore, we focus on introducing a novel routing protocol for underwater sensor networks by using Q learning based feedback mechanism. The proposed Q learning approach uses path cost, packet delivery probability and link analysis to maximize the payoff of the network. Moreover, the proposed approach is based on the opportunistic routing where the relay node is selected based on the maximum payoff. Finally, we present a simulation study to show the improved performance by using proposed approach where average network lifetime, delay and packet delivery values are obtained as 3522 rounds, 0.39s and 95.33, respectively.
Keywords:	Acoustic Networks, Q Learning, Opportunistic Routing, Underwater WSN
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	THE USE OF ONTOLOGICAL MODELING IN THE PREPARATION OF ELECTRONIC COURSES IN THE FIELD OF INFORMATION AND COMMUNICATION TECHNOLOGIES
Author:	N.ZH. SABITOVA, B.SH. RAZAKHOVA, L.U. TAIMURATOVA, Y. TIKHONOV, V. LAKHNO, R.S. SHUAKBAYEVA, R.M. BAINAZAROVA, A.A. BALEKOVA.
Abstract:	The methodological foundations of designing electronic courses in the field of information and communication technologies (ICT) based on the ontological model are outlined. This kind of modeling makes it possible to implement automated processing of ontologies for an electronic ICT textbook. A functional and informational model of the processes that form an ensemble of information technologies for automated processing of ontologies of subject areas characteristic of the field of information and communication technologies has been implemented.
Keywords:	E-Courses, E-Textbooks, Ontologies, Subject Area, UML, IDEF, DFD, Composite, Ргоху.
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	IMPACT OF ARTIFICIAL INTELLIGENCE ON THE INDIAN RETAIL INDUSTRY
Author:	Dr. DHADURYA NAIK M, Dr. SRINIVASA RAO DOKKU, VEERLA NAGAMALLESWARA, Dr. KONIKI SRINIVAS, Dr VENKATA NAGA SIVA KUMAR CHALLA, Dr M SRINIVASA NARAYANA, MUNNI VENUTHURUMILLI
Abstract:	Many of the previously laborious and labor-intensive duties connected with running a successful business are being automated by AI, which is fast revolutionising the retail industry. AI applications for retail outlets could assist firms in pricing their products by visualising the possible effects of various pricing strategies. To do so, systems gather information about other items, promotional efforts, sales numbers, and other data. The objective of the study is to know the impact of AI in Reteail industry. At Vijayawada, Andhra Pradesh, India, 145 samples from various retail businesses were taken into consideration for the study. Both primary and secondary data are used in the investigation. The study was evaluated using factor analysis. The majority of respondents are aware of the use of AI in India's retail sector, according to data studies. It has also been noted that the majority of retail establishments are using AI in their Business models. Particularly AI is useful in order processing, shipping, and inventory management in the retail industry in India.it is also identified that, most of retail owners are aware of impact of AI on their business and also they are implementing the AI techniques in their business models to meet the changing requirements of the industry.
Keywords:	Artificial Intelligence, Machine Learning, Automation, Retail, Business
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	IMPLEMENTING RESNET-50 TRANSFER LEARNING MODEL FOR DIAGNOSING OCT IMAGES FOR DETECTING AND CLASSIFYING DME ABNORMALITIES
Author:	A. GEETHA DEVI, SURYA PRASADA RAO BORRA, D. HARI KRISHNA, M. RAJ KUMAR NAIK, K. VIDYA SAGAR, LAKSHMI RAMANI BURRA
Abstract:	One of the major diseases that affect human eyes due to complications of diabetes is Diabetic Retinopathy (DR). A particular type of DR affecting the retina's central portion, called the macula, creates a vision problem. It is called Diabetic Macular Edema (DME). The blood vessels in the eye get damaged, and leaking fluid on the macula causes tissue thickening and swelling. The image may be occluded for various external and environmental reasons, which can degrade the image quality and provide a wrong diagnosis. The early diagnosis of DME is essential to avoid vision loss, and OCT images are used for prescreening because OCT is one of the non-invasive imaging modalities that can provide high-resolution retina images speedily. Several earlier research works have focused on analyzing various DR images using image processing methods, providing less prediction accuracy. This work aims to create an automatic transfer learning model for classifying DME using OCT images. ResNet-50 is created by training with 80% of the training OCT images, evaluated and validated by random, and 20% of the testing images. The reason for using the ResNet-50 model is that it is a pre-trained model using ImageNet data. The proposed ResNet is experimented with a benchmark dataset with Python for testing and validating its output. The output is compared with the other earlier methods to evaluate the performance. The comparison shows that the proposed ResNet-50 model outperforms others by obtaining 97.56% accuracy.
Keywords:	Diabetic Macula Edema, OCT Images, Transfer Learning Models, ResNet-50, Diabetic Retinopathy, Medical Image Processing.
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	REFINING THE STRUCTURE OF BAYESIAN NETWORK LEARNT WITH AGGLOMERATIVE CLUSTERING TECHNIQUE FOR PROSTATE CANCER DISEASE
Author:	NAVEEN KUMAR BHIMAGAVNI, ADILAKSHMI THONDEPU
Abstract:	Structure learning of the Bayesian Network is a two-step process, one is parameter learning and the other is finding the best structure among search space using uncertain and incomplete data. Structure learning is the most important and complex task (NP-hard problem) in estimation theory. However, existing techniques require generating all possible graphs even for a small number of random variables, and consume a large amount of space and time complexity to verify each of them. Clustering-based Structure learning can be used to learn the structure of the Bayesian network to overcome this limitation. However, the learned structure needs to be refined as and when the new data arrives and existing refinement techniques verify the relation of a node with every other node; which consumes large time complexity. In this work, we propose an algorithm that refines the structure of the Bayesian network learned using the agglomerative clustering technique using the proposed refinement algorithm. It considers only a subset of nodes (identified using Markov Assumption) for comparison of a node and thereby consumes comparatively less time complexity. Also, the Bayesian score is calculated for each candidate structure to find the best network structure.
Keywords:	Refinement Algorithm, Agglomerative Clustering, Bayesian network, Bayesian Score, Prostate Cancer, Markov Assumption
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	ENHANCED COLLABORATIVE TASK EXECUTION IN EDGE COMPUTING BASED ON BELIEF RULE GEO CLUSTERING AND PARTICLE SWARM JOINT OPTIMIZATION
Author:	UDAYAKUMAR K, RAMAMOORTHY S
Abstract:	Multi-access Edge Computing (MEC) has recently been suggested as an addition to cloud computing. MEC servers are placed close to the network's edge to reduce latency and relieve demand on cloud data centres. The cloud is more resourceful than the MEC server, which has less resource. Each MEC server cannot satisfy all computational and large data needs from user devices when operating independently. Collaborative edge computing (CEC) is a popular new concept in which devices on the edge work together by sharing data and computer resources. CEC must determine when and where each task is executed, making task offloading an important problem to address. However, it is challenging to solve task offloading in CEC because tasks can be offloaded to neighbouring devices, resulting in bandwidth contention among network flows. Most existing works do not jointly consider network flow scheduling, which can result in network congestion and inefficient completion time performance. This paper suggests a unique approach for planning, executing, and clustering MEC servers in a resource allocation model. Belief rule Geo clustering method used to cluster the MEC server based on real-time data of intensive tasks. For the simple sake, real-time monitoring application based on video surveillance is taken. The unsupervised cluster algorithm used to distribute the software components optimally among mobile devices and reduce cluster traffic in relation to the data center. The distributed MEC channel task allocation model is used to share tasks when the MEC server is busy receiving data, and a particle swarm-based joint optimization architecture is used to optimize clustered data. Extensive experiments have been performed on three datasets concerning energy utilization, delay, computational burden, throughput, and network Quality of service (QoS). The proposed technique achieved better results than benchmark solutions, which do not make a joint decision.
Keywords:	Mobile Edge Computing, Collaborative Tasks Scheduling, Resource Allocation, Geo Clustering, Particle Swarm Optimization
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	AUTOMATIC DETECTION AND CLASSIFICATION OF LEUKEMIA FROM BLOOD SMEAR IMAGE USING SENET CONVOLUTIONAL NEURAL NETWORK
Author:	SHALINI V, K. S. ANGEL VIJI
Abstract:	Leukaemia is a subtype of blood cancer that manifests itself by excessive production of abnormal blood cells. Leukemia is on the rise, and it takes more time to diagnose and treat because of the disease's high mortality rate. One such disease is leukaemia, which affects the white blood cells. Early, reliable, and safe detection of leukaemia dramatically improves treatment and survival rates. Pre-processing, segmentation, extraction of features, and the leukaemia classifier are the four components of the classification methodology. Using a hybrid Squeeze-and-Excitation Networks (SENet)-based CNN (convolutional neural network), this work can identify ALL (acute lymphocytic leukaemia), CML (chronic myeloid leukaemia), CLL (chronic lymphocytic leukaemia) and AML (acute myeloid leukaemia) in this study. The proposed SENet-CNN model detects the classification of four leukaemia subtypes, including ALL, AML, CLL, and CML, from blood smear images. The performance of the proposed SENet model is assessed using metrics like accuracy (ACC), specificity (SP), Precision, F1- score, and sensitivity (SE). The proposed model outperforms the other techniques by having a classification accuracy of 99.98%, according to experiments on a dataset of 655 images.
Keywords:	Leukemia Detection, SENet-CNN, Blood Smear Images, Deep Learning, Convolutional Neural Network, Segmentation and Classification
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	DEVELOPMENT OF NEURAL NETWORK MODELS FOR OBTAINING INFORMATION ABOUT NODULAR NEOPLASMS OF THE THYROID GLAND BASED ON ULTRASOUND IMAGES
Author:	ILYA LOZHKIN, KSENIA TSYGULEVA, KONSTANTIN ZAYTSEV, MAXIM DUNAEV4, SVETLANA ZAKHAROVA, EKATERINA TROSHINA, ALEKSANDER GARMASH
Abstract:	Analysis in ultrasound examinations of the thyroid gland often requires a large amount of training data, therefore, one of the important problems of using deep architectures in medicine is the expansion of training datasets to more significant volumes, since in most clinics the sets of such data are not large. The purpose of this study is to develop a set of neural network models for solving the problems of analyzing ultrasound images of the thyroid gland to identify nodular neoplasms, as well as to study various approaches to increasing the amount of data for the formation of training samples. According to the results of the conducted experiments, it was found that an increase in the number of the same type of images in film loops did not affect the operation of deep architectures, and therefore it was meaningless. Various approaches to the augmentation of medical image sets were investigated, and it was observed that the complication of the augmentation process of images containing enlarged areas with useful information negatively affected the quality indicators of segmentation models trained on such data. Based on the conducted research, the authors propose neural network models to solve the problems of semantic segmentation and classification for use in the field of ultrasound examinations of the thyroid gland. The results obtained allow for advancing the use of artificial intelligence methods for personalized medicine for thyroid diseases.
Keywords:	Machine Learning, Ultrasound Imaging, Detection, Segmentation, Classification.
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	THE ROLE OF TWITTER IN BUSINESS, ECONOMICS, AND FINANCE RESEARCH: A BIBLIOMETRIC ANALYSIS
Author:	DIAN KURNIANINGRUM, NUGRAHA NUGRAHA, DISMAN DISMAN, BUDI SUPRIATONO PURNOMO, MUYANI KARMAGATRI
Abstract:	The Twitter application was launched in 2006 and quickly became popular. In line with the increasing popularity of Twitter, many studies have linked the role of Twitter with business, economic and financial development. This bibliometric research was aims to provide an overview of research developments regarding the role of Twitter in business, management, accounting, economics, econometrics, and finance and provide insight into what topics have not been touched on for research. This research retrieves journal publication data from Scopus and then visualises the data to provide an overview of the journal’s contribution and performance using VOS viewer. This report will be divided into research background, bibliometric method, research methods, research results (publication structure, citation structure, and graphical analysis of publications), and research conclusion. The author and country who publish most related articles come from the United States. The publisher which publishes the most is an information system and information management publisher; the other productive publisher is the one that specialises in business, management, and marketing. Keywords that often arise beside Twitter are social media, sentiment analysis, Facebook, Covid-19, social networks, machine learning, big data, text mining, and natural language processing. The research conclusion contains key funding from bibliographic analysis and the study’s limitations. To enrich the studies, other researchers can utilise different data sources, such as the Web of Science and Google Scholar, which can also be used as references. Additional data processing, such as Cite Space and SciVal, can also be added to deepen the analyses.
Keywords:	Twitter, Bibliometric analysis, Business. Economics, Finance
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	THE INFLUENCE OF INSTAGRAM SOCIAL MEDIA ACTIVITIES AS A DIGITAL MARKETING STRATEGY IN INCREASING PURCHASES
Author:	ANGGUN CYNTHIA PUTRI, ASTARI RETNOWARDHANI
Abstract:	The development of information technology and the advancement of business innovation compel business owners to stay abreast of technological advances. Businesses across various sectors are seeking innovative approaches to support their business objectives. One such technological development is the emergence of social media networks. With the numerous conveniences offered by this new communication medium, social media users can swiftly disseminate or seek out messages and information. In Indonesia alone, as of February 2022, there were 191.4 million social media users out of a total population of 277.7 million individuals. This scenario presents an opportunity for conventional entrepreneurs to leverage technological advancements and bolster their online business for increased profitability. According to Indonesian Digital User Data, the number of social media users is recorded relatively high. XYZ, operating in the automotive sector and specializing in marketing truck and bus products, has embraced social media as a marketing platform for its offerings. Hence, this study aims to examine the impact of social media activities undertaken by XYZ company on sales performance. This research employs a quantitative approach with a target population of 521 customers familiar with Hino's Instagram social media account. The sample consists of 150 companies classified as VIP Loyal Customers of Hino products by the company. The study incorporates six interconnected variables: Social Media Marketing Activities, Brand Equity, e-WOM, Purchase Intention, Customer Satisfaction, and Customer Loyalty. The aim is to explore the influence of Social Media Marketing Activities on Purchase Intention, specifically the sales intensity of XYZ company. The results of hypothesis testing conducted using SmartPLS 4.0 software yielded six significant and accepted hypotheses. The findings indicate that Social Media Marketing Activities positively impact brand equity, which, in turn, positively influences e-WOM. Furthermore, e-WOM positively affects purchase intention, which, in turn, positively impacts customer statistics. Additionally, customer satisfaction influences customer loyalty, and social media marketing activities positively influence customer loyalty. Consequently, the research outcomes can serve as valuable insights for augmenting Hino product purchases at XYZ company.
Keywords:	Social Media, Social Media Marketing Activities, Brand Equity, ewom, Purchase Intention, Customer Satisfaction, Customer Loyalty
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	DESIGN OF MUTATION OPERATORS FOR TESTING USING PARALLEL GENETIC ALGORITHM FOR OPEN-SOURCE ENVIRONMENTS
Author:	SANDDEP KADAM, T. SRINIVASARAO
Abstract:	Specification-based testing approaches create test data without having any prior knowledge of the program's structure. However, the quality of this test data isn't always reliable enough to catch errors when non-functional modifications are made to the software. We offer a novel technique that combines formal requirements and the evolutionary algorithm to successfully produce test data. In this technique, Parallel Genetic Algorithm (PGA) rewrites formal requirements in order to create input values that kill as many mutants of the target programmed as feasible. To explain how the approach works, two famous instances are offered. The results suggest that the proposed technique may successfully produce test cases to eliminate programmed mutants, resulting in improved software maintenance.
Keywords:	Model-Based Testing; Fault Localization; Search-Based Algorithm; Automatic Test-Case Generation; Mutation-Based Testing
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	LEVEL OF STUDENT SATISFACTION WITH NEW BINUSMAYA: MEASURING AND ANALYZING USING THE END USER COMPUTING SATISFACTION (EUCS) FRAMEWORK
Author:	RAYHAN FADLI ROBBY , TUGA MAURITSIUS
Abstract:	Binus University has used Binusmaya as a learning tool since 2001. According to the official Binus website on December 25 2019, the switch from Binusmaya to New Binusmaya was made to make this platform better and more useful. Development is carried out by adding the latest features and improving the appearance design. This is done in order to make students feel comfortable in finding, exploring, giving, and exchanging important information about knowledge and lecture activities. Basically new binusmaya is a refinement of binusmaya. Some of the differences between binusmaya and new binusmaya are the additional features in the new binusmaya, which are features that function to display information about student progress for all courses taken in a particular semester or period. This article discusses the level of student satisfaction with the website version of the new Binusmaya and the factors that influence student satisfaction with the website version of the new Binusmaya. The framework used is End User Computing Satisfaction (EUCS) which functions to determine the level of satisfaction of application users by comparing expectations and reality from information systems. The population to be used in this study were active students at Binus University with a sample of undergraduate students consisting of 8 faculties with a total of 420 students as respondents. Questionnaire distribution starts from 26 September 2022 – 8 November 2022. The results show that the content variables (X1) and accuracy (X2) have no significant effect on user satisfaction (Y) while the format variables (X3), ease of use (X4) and timeliness ( X5) has a significant effect on user satisfaction (Y).
Keywords:	New Binusmaya, Learning Management system, EUCS, E-learning, User Satisfaction
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	DETERMINING HEUTAGOGY DESIGN ELEMENTS FOR ONLINE LEARNING MODEL USING FUZZY DELPHI METHODS
Author:	FAIRUZZA HAIRI, SITI NURUL MAHFUZAH MOHAMAD, SAHRUDIN SAAD, IBRAHIM AHMAD
Abstract:	Teaching and learning in computer application courses are being improved by educators as technology continues to change the face of education. For the purpose to promote student participation and self-directed learning, this study investigates how the heutagogical online learning paradigm might be adapted to computer application instructions. The Heutagogy Online Learning Model (HeuLEM) is a model that this research proposes based on heutagogy principles, including the six design elements of heutagogy. It enables learners to take control of their educational process, make knowledgeable choices, and develop critical skills for the digital age. Incorporating self-determined learning, reflection, strong interpersonal skills, and collaborative online tools, instructors can design an engaging and dynamic online learning environment that fosters students' digital collaboration abilities. HeuLeM comprises the theoretical underpinnings of heutagogy and offers helpful implementation tactics for the model in computer application courses. This study is to identify whether the six heutagogy design elements can be adapted to Heutagogy Online Learning Model to enhance the digital collaboration skills of students. A total of 12 experts were selected to analyze the fuzziness consensus of experts. All of the data obtained will be analyzed using the Fuzzy Delphi method. The findings show that all the heutagogy design elements are suitable to be adapted Heutagogy Online Learning Model to enhance the digital collaboration skills of students. The six heutagogy elements that explore, create, connect, share, reflect, and collaborate have reached unity with the threshold value of d ≤ 0.2 and the percentage of the expert group is more than 75%. Therefore, based on the consensus agreements, the development of Heutagogy Online Learning Model for enhancing the digital collaboration skills of students should be incorporated with the elements of heutagogy.
Keywords:	Heutagogy; Collaboration Skills; Online Learning Model; Fuzzy Delphi; Heutagogy Design
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	UTILISING A DYNAMIC APPROACH TO REDUCE TARDINESS FOR SCHEDULING ISSUES WITH DISTINCT DUE DATES AND JOB BLOCKING TECHNIQUE TO REDUCE COST OF TARDINESS AND EARLY ARRIVAL
Author:	E.JANAKI, A.MOHAMED ISMAIL, SINI RAHUMAN
Abstract:	The goal of this work is to reduce the overall earliness cost when scheduling independent jobs with varied due dates on a single machine. Dynamic programming for a single machine and Johnson's method based approach for a flow shop are the two precise solutions suggested for the problem. In the manufacturing and service industries, sequencing and scheduling is the critical part of decision-making. Effective sequencing and scheduling has become a requirement for market survival in today's competitive economy. Companies must adhere to shipping deadlines that have been promised to clients. In this paper, we will discuss about different types of machines for different types of jobs through Phase –I ,Phase–IIand Phase III. The objective ofthis paper to minimize the Tardiness by using Dynamic approach and job blocking method.
Keywords:	Single Machine, Flow Shop, Earliness, Tardiness, Dynamic Approach, Blocking Jobs
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	ANALYSIS OF QOS ROUTING IN MOBILE AD HOC NETWORKS USING ANTHOCNET, SHUFFLED FROG LEAP, FIREFLY AND LION OPTIMIZATION ALGORITHMS
Author:	K.SUMATHI, DR.J.K.KANIMOZHI
Abstract:	A Mobile Ad Hoc Network (MANET) is a network of wireless nodes. Base stations, infrastructure, and infrastructure are not required for a mobile ad hoc network. MANET is a network that is only available temporarily. MANET technology is used in military and search and rescue operations. Quality of Service (QoS) is a key term for the overall optimization of network resources. Mobile Adhoc networks are well-known for their self-organization and autonomy. QoS-based routing over MANET necessitates an adaptive and fast path search solution. Optimization algorithm-based techniques, such as ant colony optimization (ACO) algorithms, shuffled frog leaping algorithm, Lion Optimization Algorithm (LOA), and Firefly Algorithm (FF), have proven to be effective in developing routing algorithms for mobile ad hoc networks. Then, an optimal path is computed using hybrid ACO based routing, which is an efficient routing scheme based on foraging ant behaviour. The proposed hybrid algorithm outperforms the Anthocnet and shuffled frog leap, firefly, and lion optimization algorithms in simulation (LOA). The proposed SFLAO algorithm also outperforms the standalone Anthocnet and LOA, Shuffled frog leap, and Firefly algorithms in terms of QoS performance metrics.
Keywords:	Mobile Ad Hoc Network, Routing Protocol, Quality Of Service, Lion Optimization Algorithm, Ant Colony Optimization, Shuffled Frog Leaping Algorithm
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	FACIAL RECOGNITION MODEL AS EMPLOYEE ATTENDANCE USING MODERN DEEP LEARNING WITH CONVOLUTIONAL NEURAL NETWORK METHOD
Author:	HALIM MAULANA, FERDY RIZA , AL-KHOWARIZMI
Abstract:	Development of technology is developing very quickly, thus providing many benefits, especially in the field of Information Technology. Demand for services with the use of technology is increasingly needed by various industrial fields, especially the impact generated by Covid-19 has resulted in the application of technology being required in various fields to minimize physical contact. The conventional attendance process by touching the attendance device/devices can be replaced by using face recognition technology. Face Recognition technology uses Deep Learning discussion. Using this technology, without needing to touch employees, they can record attendance by looking into the camera. This study uses the Convolutional Neural Network (CNN) algorithm. The programming language used in this program is Python. The process of making this application with the stages of making Face Recognition, namely image acquisition, preprocessing, extraction, classification, and identification of image data. The dataset is separated into 3 stages of data, namely train data, validation data, and test data.
Keywords:	Deep Learning, Machine Learning, TensorFlow, Python, Facial Recognition
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	COLLABORATIVE ANT COLONY OPTIMIZATION-ASSISTED SUPPORT VECTOR MACHINE FOR ACCURATE COTTON LEAF DISEASE CLASSIFICATION AND YIELD PREDICTION
Author:	S.GOVINDASAMY, D.JAYARAJ
Abstract:	Cotton leaf diseases pose a significant threat to the economic viability of cotton farming, leading to substantial yield losses. However, the lack of a reliable classification system and accurate yield prediction based on disease profiles hinders effective disease management and resource allocation. This research tackles these issues by recommending a novel method for predicting cotton production and classifying cotton leaf diseases using a combination of Ant Colony Optimization (ACO) and Support Vector Machines (SVM). The research objectives include developing a customized Collaborative ACO (CACO) algorithm for feature selection, implementing an SVM model for disease classification, and integrating the CACO-SVM framework for accurate disease identification and yield prediction. The suggested method will be tested on a labelled dataset of cotton leaf samples annotated with disease information. The importance of this study rests in its ability to allocate resources to cotton growing better and improve disease management tactics. Farmers can adopt targeted management practices, reduce production costs, and mitigate the environmental impact of generalized treatment approaches by accurately classifying cotton leaf diseases and predicting crop yields. The outcomes of this study are expected to contribute to improved agricultural practices, increased profitability for cotton farmers, and enhanced sustainability in cotton production. The proposed CACO-SVM framework has the potential to revolutionize disease identification and yield prediction in cotton farming, empowering farmers to make informed decisions and achieve higher crop management efficiency.
Keywords:	Classification, Ant Colony Optimization, Support Vector Machines, Yield Prediction, And Cotton Leaf Disease
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	HYB-ICNN: HYBRID INCEPTION CONVOLUTIONAL NEURAL NETWORK FOR FEATURE EXTRACTION AND HYPERSPECTRAL IMAGE CLASSIFICATION
Author:	EASALA RAVI KONDAL, SOUBHAGYA SANKAR BARPANDA
Abstract:	The remote sensing research community has devoted particular focus to the classification of hyperspectral images (HSI). To improve the classification accuracy of hyperspectral images (HSI), deep learning-based technology has been proposed. However, it remains a challenging obstacle to achieve satisfactory classification accuracy with insufficient data for training. Therefore, a more effective neural network design needs to be devised in order to boost the effectiveness of the HSI classification function. To address this concern, this letter provides a novel Hybrid-Inception CNN (Hyb-ICNN) framework for dynamically obtaining characteristics by laying inception components in the model which can acquire better accurate properties with smaller training samples by employing volatile spatial dimension convolutional filters and dynamic CNN framework. The outcomes of the experiments show that the presented model can boost classification performance by adaptively changing the network structure. The experiments are carried out on both the new data sets and the publicly accessible benchmark data sets to expose the efficiency and durability of the proposed model. The proposed Hybrid-Inception CNN model has achieved accuracies of 80.79% on the AH1 (Ahmedabad-1) dataset, 87.98% on the AH2 (Ahmedabad-2) dataset, 99.99% on the PU (Pavia University), 99.99% on the SA (Salinas), and 99.92% on the IP (Indian Pines) dataset. Empirically, it has been demonstrated that the presented model succeeds over the remaining state-of-the-art approaches in terms of classification accuracy.
Keywords:	Hyperspectral Image, Inception Network, Convolutional Neural Network, Feature Extraction, Classification
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	INDUSTRY-UNIVERSITY ENGAGEMENT, INFLUENCE, AND BENEFITS: COMPUTING JOB MARKET IN KUWAIT
Author:	EIMAN TAMAH AL-SHAMMARI
Abstract:	Due to increasing competition among the IT sector in Kuwait it has become difficult for fresh graduates to find a steady and inspiring positions at private organizations. Employer’s demand updated cognitive skills which the universities have failed to provide. Thus, the new job trend has encouraged a strategic collaboration between industries and academic schools. This study attempts to reconfirm the same. Through the set of questionnaires, we intend to find the opinions of fresh graduates regarding their employment scope. Also, whether companies’ strong associations can influence employment opportunity. To establish reliability, Cronbach's coefficient alpha (α) was employed to analyse the measure's internal consistency. The reliability of the questionnaires was assessed using the partial least squares regression (PLS) approach. To assess the data and validate the hypotheses, a structural equation model (SEM) was used. Out of all the 4-hypothesis postulated. Our data also indicated that the level of collaboration between Kuwaiti Universities and industries and corporations is not so strong. In conclusion, focus needs to be on increasing collaboration between university and corporates sector. This will have a positive influence on Opinion and hiring strategy.
Keywords:	Academia–Industry, Higher Education, School-Industry Collaboration, Job Market, Computing, IT, Employability Skills
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	NEW APPROACH FOR RESOLUTION OF MAXWELL EQUATIONS USING GALERKIN METHOD AND DIRICHLET CONDITIONS IN FINITE ELEMENT ANALYSIS
Author:	WALID EL FEZZANI, OSAMA YASEEN M. AL-RAWI, WISAM SUBHI AL-DAYYENI
Abstract:	The Galerkin method combined with the Dirichlet conditions provides a powerful way of solving the Maxwell equations. This approach allows for a detailed description of the behavior of the electric and magnetic fields in the problem domain and can provide accurate solutions for complex engineering problems. The method can be used in conjunction with Finite Element Analysis to solve various types of problems, such as those involving loads, force, displacement, vibration, and fluid flow. The use of the Dirichlet conditions to impose boundary conditions on the system also ensures accuracy in the results obtained. This helps engineers to better understand the physical behavior of the system, assess any discrepancies in the data, and make more informed decisions about the overall design. FEA combined with Dirichlet conditions offer an efficient and accurate way of predicting design behavior and allow engineers to make changes to the design quickly and efficiently. The proposed numerical solution for MAXWELL equations using the GALERKIN method and DIRICHLET conditions is presented by a Finite Element Analysis mesh distribution to calculate the flux and the magnetic field into the air gap of the electrical motor.
Keywords:	Finite Element Analysis, Galerkin, Dirichlet, Maxwell Equations, Electric Motors
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text

Title:	NEXT GENERATION SD-WAN WITH IDPS
Author:	JOSEPH NG POH SOON, PHAN KOO YUEN, FADHILAH NUR RANIA, CHAN JIA YI, FONG JUN YIP, NG EE SERN, SITI ZUBAIDA BINTI ZULKIFLI
Abstract:	The Wide Area Network and firewall are the current networking technologies that are widely used in most enterprises. However, due to the rapid development of new technologies, there is a rising demand for a new efficient networking solution that could overcome the current technologies' limitations and cater to the end users' new requirements. This research integrated Software Defined Wide Area Networks and Intrusion Detection and Prevention Systems. This paper aims to conduct a systematic review where the effectiveness and benefits of the new technologies are investigated as a proposed solution. that optimized the network to prevent the trombone effect that degrades network performance while recognising malicious activities and preventing attacks. It can also easily detect and alert the network administrator. Combining both technologies is the proposed solution that will enable network performance and security improvements, resulting in an efficient network solution.
Keywords:	Firewall, WAN, SD-WAN, IDPS, Network
Source:	Journal of Theoretical and Applied Information Technology 15^th August 2023 -- Vol. 101. No. 15-- 2023
Full Text