		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
January 2023 | Vol. 101 No.2

Title:	THE ROLE OF MOTIVATIONAL THEORIES IN THE SUCCESS OF CROWDSOURCING ENGAGEMENT MODELS: A REVIEW
Author:	HASAN HUMAYUN, MASITAH GHAZALI, MUHAMMAD NOMAN MALIK
Abstract:	A plethora of research has been conducted on motivational theories in various fields including medical sciences, business and management, physiology, and sociology, especially in the natural sciences field. Motivational theories are considered a key to motivating the crowd over the internet to participate in the assigned tasks over online platforms commonly known as online crowdsourcing. However, research regarding the review of the theories discussed is scarce. Therefore, this literature review focuses to identify the motivational theories in literature over the last decade and mapping these theories onto the engagement models of crowdsourcing. Based on a review of 91 papers from the natural science domain, we identified 36 motivational theories and mapped the identified theories over crowdsourcing models of engagement. The analysis of the study identified the popular theories among the researchers as well as the new and nascent theories practiced in Crowdsourcing, from 2010 to 2021. Similarly, the mapping helped to identify the nature of the contribution and the theories' importance. The literature review help to understand the recent trends to motivate participants using motivational theories and help identify trends and possibilities for future research.
Keywords:	Motivation; Theories; Crowdsourcing; Engagement; Models; Literature Review.
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	PARALLEL K-MEANS FOR BIG DATA: ON ENHANCING ITS CLUSTER METRICS AND PATTERNS
Author:	VERONICA S. MOERTINI, LIPTIA VENICA
Abstract:	K-Means clustering algorithm has been enhanced based on MapReduce such that it works in distributed Hadoop cluster for clustering big data. We found that the existing algorithm have not included techniques for computing the cluster metrics necessary for evaluating the quality of clusters and finding interesting patterns. This research adds this capability. Few metrics are computed in every iteration of k-Means in the Hadoops Reduce function such that when it is converged, the metrics are ready to be evaluated. We have implemented the proposed parallel k-Means and the experiments results show that the proposed metrics are useful for selecting clusters and finding interesting patterns.
Keywords:	Clustering Big Data, Parallel k-Means, Hadoop MapReduce
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	AIR QUALITY CONTROL USING HOTELLING DOUBLE BOOTSTRAP CONTROL CHART WITH VECTOR AUTOREGRESSIVE MODEL APPROACH
Author:	JAUHARIN INSIYAH, SUCI ASTUTIK, LOEKITO ADI SOEHONO
Abstract:	This study aims to find a sensitive control chart to shifts in the control process that induces a correlation among time series observations, known as autocorrelation. Hotelling, a popular multivariate chart, is no longer sensitive to detect small and moderate mean shifts derived from the autocorrelation process. Therefore, this study uses Vector Autoregressive Model (VAR) residuals to build Hotelling control charts. To improve the chart Double Bootstrap method is used to construct a sensitive control limit because the assumptions of the Hotelling are not fulfilled. Violation of assumptions results in the analysis being inappropriate. The proposed control chart is used for air quality control in Surabaya with the characteristics quality of PM 2.5, PM 10, and CO, which are correlated with each other. The proposed control chart’s performance is compared with the single Bootstrap control chart by Average Run Length (ARL) value at different numbers of observations. The results show that The Proposed based on residual VAR with Double Bootstrap is more sensitive than the single Bootstrap to detect out-of-control on all shifts and at observations. Thus, the proposed control chart can be a way to minimize errors in controlling air quality.
Keywords:	Air Quality Control, Double Boostrap, Multivariate Control Chart, Hotelling, Vector Autoregressive (VAR)
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	PROPOSED APPROACH FOR ACADEMIC PAPER RANKING BASED ON BIG DATA AND GRAPH ANALYTICS
Author:	NAGWA YASEEN HEGAZY , MOHAMED HELMLY KHAFAGY , AYMAN E.KHDER
Abstract:	The outburst growth of technology in the academic environment and the widespread use of digital libraries have generated big scholarly data. Ranking and measuring the impact of academic papers grants higher importance to the academic environment that is required for promotions, hiring, awards, grants, scholarships, and ranking university procedures. Google Scholar ranking depends mainly on the citation count of academic papers; therefore, some papers are ranked low even if they are qualified papers. Identifying the most important articles in the field is considered a critical issue for researchers, journals, and academic institutions. The goal of this study is to create a ranking system for big scholarly data (RBSD) that integrates network analysis based on graph analytics, citation analysis, and similarity between papers. The proposed model ranks papers based on the paper citation network to get the central papers. It also ranks authors to identify the top authors in the computer science citation network and analyzes the similarity between academic papers to get the relevancy between papers. A new methodology is proposed to rank papers based on a weighted score that considers paper information, author information, and publication venue information. The proposed model also considers the complex relationship between papers, overcoming the limitations of other ranking systems that rely only on the traditional PageRank algorithm. To produce a more accurate ranking system, our suggested model excludes authors self-citation and collaboration citations, which are often used by authors to increase their citation count. To evaluate the RBSD model, four real-world datasets were used: ACM, MAG, DBLP, and Scopus Elsevier, for publication venue information. The proposed model was applied to 2,092,356 papers, with 8,024,869 citations. This was implemented using Apache Spark Graphx to accelerate the execution time for graph analysis and to explore the nature of scholarly data. The experimental results show that our proposed model outperformed the Google Scholar Ranking procedure based citation count and returns reasonable results.
Keywords:	Scholarly Data, Big Data, Graph Theory, Citation Analysis, Ranking Systems, Bibliographic Coupling, Co-Citations.
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	EXAMINING THE FACTORS FOR NON-COMPLIANCE OF SAUDI HEALTH ORGANIZATIONS FOR E-HEALTH SECURITY AND PRIVACY
Author:	KHALIL IBRAHIM ALMUWAIL, ABDULAZIZ SAAD ALBARRAK, MUHAMMAD NASIR MUMTAZ BHUTTA, HEIDER A. M. WAHSHEH
Abstract:	Recently, there has been increasing concern about the security of patient data and the impact of Saudi health organizations non-compliance with privacy and electronic health security regulations. However, there is a limited number of studies related to security compliances in Saudi healthcare organizations. This research aims to study the challenges of security compliance standards in hospitals for patient data security and privacy in the Saudi Arabian healthcare sector. Healthcare facilities of all sizes need help in maintaining security rule compliance. Analysis of factors influencing complete adherence to security rule compliance has begun as part of the non-compliance study. Various theoretical frameworks and conceptual models have been used in previous studies to help small and medium healthcare facilities comply with security compliance successfully. In addition, there is a demand to utilize security compliance frameworks in hospitals, and national security standards are essential for protecting patient data. Therefore, this study intends to investigate the factors that lead to non-compliance with E-Health security and privacy in Saudi health organizations. A set of hypotheses were developed to achieve the studys goal. Furthermore, the study follows a quantitative approach to evaluate the proposed model and hypotheses. Statistical Package for Social Sciences (SPSS) is employed to analyze the users responses, and the IBM analysis of moment structures (AMOS) is used to validate the research model. The findings suggest that understanding management support, security awareness, security culture, and computer self-efficacy is essential to ensure compliance with the security rules.
Keywords:	Security compliance, E-health security, data privacy, data protection, security non-compliance
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	A PROPOSED GAMIFICATION FRAMEWORK USING SENTIMENT ANALYSIS AND FUZZY LOGIC IN HIGHER EDUCATION
Author:	GHADA KHAIRY, ABEER M. SAAD, SALEM ALKHALAF, MOHAMED A. AMASHA
Abstract:	The traditional methods used to analyze and collect student feedback are not scalable, so determining levels of student satisfaction is difficult and entails various challenges. This paper aims to understand students sentiments about the use of gamification in higher education. First, we measured student satisfaction with sentiment class. We observed that there is a direct relationship between the sentiment scores of the Senti WordNet lexicon (SWN) and student satisfaction level. If the sentiment score of the SWN lexicon increases, then student satisfaction also increases. The student satisfaction level was 81% for the SSAGS dataset. Furthermore, when using SVM, NB, and DT classifiers, we found that some aspects yield high results because students opinions are positive, and their satisfaction levels are higher. For example, the accuracy of the motivation aspect is equivalent to 100% with the SVM and DT classifiers. Additionally, the accuracy of the clarity aspect and the improvement aspect is equivalent to 92.5% with the NB classifier. Second, the SSAGS dataset was evaluated, and this dataset comprises two different experiments, one using the SWN lexicon and the other using SVM, NB, and DT classifiers. Finally, the results showed high accuracy and high recall in the process of analyzing student opinions to determine the level of student satisfaction.
Keywords:	Sentiment Analysis, Gamification, Fuzzy Logic, Higher Education
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	APPLICATION OF INFORMATION TECHNOLOGIES AND METHODS FOR PROCESSING BIG DATA TO THE MANAGEMENT OF THE EDUCATIONAL PROCESS DURING THE PANDEMIC
Author:	MUKHIYADIN A., MAKHAZHANOVA U., SERIKBAYEVA S., KASSEKEYEVA A., MURATOVA G., KARAUYLBAYEV S., MURATKHAN R., KENZHEBAY A.
Abstract:	The spread of coronavirus infection has changed all spheres of human life, forcing a new look at traditional forms of activity. Education as the most important social sphere has taken the blow of the pandemic transformation, resulting in a significant spread of distance forms of interaction between the teacher and the learner, involving the reliance on digital learning tools. Due to the need for participants in the educational process in these conditions to operate with large amounts of data and process information electronically, the use of the potential of artificial intelligence in education. The article deals with new opportunities and problems of integration of education and artificial intelligence under conditions of Covid-19 pandemic. The object of the research is education in the epoch of Covid19. The article describes fundamental changes, which took place in the educational system during the pandemic period, outlines promising directions of artificial intelligence usage in the modern educational process and corresponding problems. The necessity of further research of possibilities of integration of education and big data under conditions of acceleration of processes of digitalization of society is marked. COVID-19 epidemic caused a great number of human casualties and destruction of economic, social, public and health systems all over the world. Combating such an epidemic requires an understanding of its characteristics and behavior, which can be determined by collecting and analyzing relevant big data. Big data analysis tools play a vital role in generating the knowledge necessary for decision-making and precautionary measures. However, because of the vast amount of COVID-19 data available from various sources, there is a need to review the role of big data analysis in controlling the spread of COVID-19 by presenting the main challenges and directions of COVID-19 data. analysis, as well as providing a framework for relevant existing applications and research to facilitate future research on COVID-19 analysis. Therefore, in this paper, we conduct a literature review to highlight the contributions of several studies in the field of COVID-19-based big data analysis.
Keywords:	COVID-19, Big Data, Relationship Between Traits, Social Data, Regression Analysis, Multinomial Logistic Regression, Secondary School Students, Educational Activity
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	WHO ARE THE MESSENGERS ON SOCIAL MEDIA THAT THE YOUNG GENERATION TRUST?
Author:	ANDRE PRANATA , MUSTIKA JATI , LA MANI
Abstract:	Trust in the traditional media has declined. Likewise, social media, which has become an alternative source of information has significantly lost the trust of its users even though it has been an integrated part of social life. In this study, social media position was a medium/tool to disseminating information to users. Then this information is interpreted based on the perceptions of each individual. So what information on social media does the young generation believe?. This study aims to answer this question. An online survey was conducted using cluster random sampling. The respondents were 365 young people of social media users living in the area of Greater Jakarta. Research result show that Religious leaders became the most dominant opinion leader compared to other opinion leaders, followed by public officers who have authority being among the most trusted source of information. Information from the outer circle of social media users is more trusted than that from the inner circle. Furthermore, popularity of mass media remains a reliable source of information. The actors behind information proliferation in social media significantly influence perception and trust in the information received by the users.
Keywords:	Information; Social Media; Trust; Young Generation, Messenger
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	HAS COVID-19 AFFECTED SOFTWARE USABILITY: MOBILE ACCOUNTING SYSTEM AS A CASE
Author:	YOUSEF AWWAD DARAGHMI, BAHAA YAHYA, EMAN YASER DARAGHMI
Abstract:	The importance of electronic systems has increased due to COVID-19 because of the mobility constraints which stimulates businesses to look for remote work supporting systems. So, businesses either rapidly adopted off-the shelf software or demanded the development of new software solutions. This causes usability concerns including new difficulties to businesses because of the low usability of the off-the shelf systems that were not designed to address the challenges during the pandemic. Also, the development of new software usually requires much time and may not produce usable software if all requirements are not sufficiently addressed. Furthermore, the factors affecting usability after the COVID-19 have not been identified in related studies. Therefore, this research empirically investigates the usability of software by developing a mobile accounting system and conducting qualitative analyses to evaluate the system and identify the usability factors in the post COVID-19 era. This research contributes the Rapid Application Participatory Development (RAPD) method which is used to develop the system because this method enables rapid development, sufficient requirements elicitation by allowing users to participate in the design process, and usability testing during the development. The results show that the RAPD method can be used to develop usable software and mobile applications. Also, in addition to the traditional usability factors, COVID-19 has created new usability factors including remote work, user experience, security, privacy, artificial intelligence and internet speed.
Keywords:	Mobile Accounting System, COVID19, Rapid Application Participatory Development, Usability testing.
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	COMPARISON OF CLASSIFICATION METHODS AND CLUSTERING HYBRID DEEP NEURAL NETWORK DETECTION OF SENSITIVE INGREDIENTS IN FOOD PRODUCTS
Author:	T.H.F HARUMY, D.S GINTING, F.Y MANIK
Abstract:	Cases of Death of Indonesian People Due to Food Poisoning, Especially Packaged Foodstuffs are Assessed as High Enough. According to BPOM data, cases of food poisoning are more than 2000 cases per year. This is due to the lack of literacy received by the community about the composition of ingredients in food products. So that innovation is needed with an in-depth analysis of the design of a smart system that can be used by the public to identify certain compositions or ingredients shown in the composition table of a product. Before designing a smart system, an in-depth analysis will be carried out using the Hybrid Deep Neural Network method and comparing it with other methods in order to get the best method that can be implemented in the smart system later. This study aims to classify and cluster product image data and food composition data by developing a hybrid deep neural network and comparing with other methods, namely KNN, Tree, SVM, and linear regression as well as clustering using Hierarchical clustering and K-Means methods. a system designed from a combination of these methods can provide accurate, effective and efficient detection results. The stages of the research method are (1) Observation and Collection of Image Data and categorical composition of Food Composition, Nutritional Value, and Characteristics of food, (2) propocesing categorical datasets (3) Embedded Image data (4) Clustering data using Hierarchical Clustering and K-Means ( 5) classification of image data using Deep Neural Network (6) classification of image data using KNN, Tree, SVM, and linear regression (7) Training and testing models (8) Comparing and Evaluation Models. From the results of the analysis, it was found that for Categorical data, the RMSE value of the KNN model was 0.085, the MSE value was 0.007, the MAE value was 0.068 and the R2 value was 0.158. Furthermore, the Tree model has the RMSE value of 0.118, the MSE value is 0.014, the MAE value is 0.091 and the R2 value is 0.608. Furthermore, the SVM model, the RMSE value is 0.091, the MSE value is 0.008, the MAE value is 0.078 and the R2 value is 0.044. Furthermore, the Neural Network RMSE value is 0.117, the MSE value is 0.014, the MAE value is 0.089 and the R2 value is 0.574. and the last is linear regression, the RMSE value is 0.086, the MSE value is 0.007, the MAE value is 0.067 and R2 is 0.153. so that the best model in this case is KNN which is 0.915 or 91, 5% . Furthermore, for Categorical, the results are that the KNN precision value is 0.457, then the Tree precision value is 0.390, for the SVM method the precision value is 0.311, the Precision Neural Network value is 0.387, the nave Bayes precision value is 0.336 and the logistic Regression is 0.349 and the highest precision value is for this case is KNN. Furthermore, for Hierarchy clustering, very good results were found with the number of Clusters as much as 2 clusters, namely Sweet and Salty, and more Salty clusters were seen and for hypertension disease, the highest calorie content of sensitive ingredients was Fat and Salt 36.56%.
Keywords:	Deep Neural Network, Sensitive Ingredients, Food Products
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	DIVERSITY ENHANCED ADAPTIVE CORRELATION CONNECTED CLUSTERING OF LONG TAIL ITEMS
Author:	SOANPET SREE LAKSHMI, T. ADILAKSHMI, BAKSHI ABHINITH
Abstract:	Recommender systems play a very important role in driving businesses. They recommend a set of items to the user which have a higher chance of getting consumed. The primary issue addressed in this work is to generate a recommendation list with items belonging to various categories so that the user can explore his different interests. The proposed method studies the diversity of the recommended list by enhancing the adaptive clustering method. In this approach, the dataset is partitioned into 3 sets namely the head part, the mid part, and the tail part . Then, different sets of methods are used to improve the diversity of the recommendation list. Popular items are extracted from the head part, items in the mid part are extracted using rating-based clustering method and the items in the tail part are extracted using correlation clustering-based method, thereby improving the diversity of the recommendation list.
Keywords:	Recommender Systems, Diversity, Clustering, Adaptive, Long Tail.
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	NOISE-ROBUST IN THE BABY CRY TRANSLATOR USING RECURRENT NEURAL NETWORK MODELING
Author:	MEDHANITA DEWI RENANTI, AGUS BUONO, KARLISA PRIANDANA, SONY HARTONO WIJAYA
Abstract:	The development of baby cry translators is still uncommon in Indonesia. Hence, this research aims to improve the Madsaz application as a noise-robust baby cry translator. Despite its success in translating the Dunstan Baby Language version of baby cry, the Madsaz application encounters a problem of decreased accuracy by up to 30% due to noise. Therefore, the objective of this research is to solve this problem using recurrent neural networks as deep learning modeling with the input of representative feature extraction. This modeling can classify and resolve noise in the dataset. This research utilizes an architecture modification of recurrent neural networks, i.e., the gated recurrent unit (GRU) and long short-term memory network (LSTM). This research also employs the Mel-Frequency Cepstrum Coefficient (MFCC) as a feature extraction method and the Dunstan Baby Language version of baby cry as the dataset. An experiment was carried out in two scenarios, namely input data without noise and input data with noise. The results show that the accuracy levels of the GRU and LSTM methods are 94% and 91%, respectively, on data without noise. On the other hand, the accuracy of data with noise is decreased by 5%, from 94% to 89%, in the GRU method, but decreased by 34%, from 91% to 57%, in the LSTM method. Hence, this finding indicates that the GRU method is more noise-robust, specifically against 5 to 20 dB of noise, compared to the LSTM method. In terms of effectiveness, the GRU is equivalent to the LSTM. However, the GRU method has more computational efficiency due to its simple network structure and fewer trained parameters, making it ideal for situations with small amounts of data and present noise.
Keywords:	Baby cry translator, GRU, LSTM; MFCC; Noise-robust; Recurrent neural network
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	THE USE OF VISUAL HUMOR ON TRAVEL COMPANIES INSTAGRAM : SEMIOTIC PERSPECTIVE
Author:	ANGGA PRIMADIPTA, LA MANI, ROBY HERVINDO, PRAWIRA AMADEUS MAYER, RICARDO INDRA
Abstract:	Travel companies are struggling using social media for the right implementation, particularly in formulating engagement-based visual messaging strategies. However, making interesting posts can generate strong engagement and lead to a more positive public view of the brand. Humor is one of the weapons used by social media users, given its ability to develop social interactions. However, how humor works in social media is not well understood, especially visually because different life experiences create different perceptions to make meaning of it. This study wants to make humor a symbolic source and adopts a combined semiotic content-analysis to identify the visual content and its symbolic meaning in tourism posts of various travel companies. 20 posts from 2 travel company brands with tourism content were collected. The results show that there are 5 types of humorous content and 4 types of symbolic meaning generated by the content. Research content is product related but not hard selling. This research advances tourism literacy, humor theory, and provides a conceptual map to travel companies in developing social media through visual humor content to increase engagement with consumers.
Keywords:	Humor Content, Social Media Marketing, Travel , Visual Message Strategic
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	EVALUATION OF THE EFFICIENCY OF BUSINESS PROCESS MANAGEMENT AS AN ELEMENT OF AN AUTOMATED QUALITY MANAGEMENT SYSTEM FOR AN ENTERPRISE OF THE REPUBLIC OF KAZAKHSTAN
Author:	TOIBAYEVA SH.D., UTEPBERGENOV I.T., BODESOVA А.E.
Abstract:	This research is devoted to the development of innovative technology for automation of the quality management system of enterprise in Kazakhstan and its adaptation to the management system of enterprise. This paper deals with quality, as an important strategic tool in business. System effectiveness evaluation of quality management enterprises is of a great importance connected with the formation of rational decisions in the management of quality management systems including specificity of quality indicators, multi-level system, necessity to choose the optimal number of performance indicators and system status evaluation. The objective and relevance of this research is connected with the need to: 1) solve the problems of quality management in the digital economy, following from the relevant National programs of the Government of the Republic of Kazakhstan, which are important at this step of in-depth scientific research; 2) guarantee the competitiveness of domestic enterprises with high quality requirements for products and services; 3) improve the efficiency of automated quality management systems; 4) saving resources (human and timing) in data processing. The method and model of automated enterprise quality management and intelligent automated system of quality management of enterprise integrated with ruling MICS subsystems (Management Information and Control System) are offered allowing to automate QMS implementation and support processes and increasing the validity, efficiency and effectiveness of management decisions by automated a number of functions of decision makers and personnel. This project was supported by a grant from the Ministry of Education and Science of the Republic of Kazakhstan (Zhas Galym project No. AR 13268939 Research and development of digital technology to provide consistency in the media of normative documents of the quality management system).
Keywords:	Quality Management System, Automated Business Processes, Quality Management, Efficiency Evaluation, Fuzzy Logic.
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	EVALUATION OF MACHINE LEARNING TECHNIQUES FOR ANOMALY DETECTION ON HOURLY BASIS KPI
Author:	ANDIKA HAIRUMAN, GEDE PUTRA KUSUMA
Abstract:	The most common used method for anomaly detection in the mobile radio network is using the fixed threshold on hourly and daily basis key performance indicator (KPI) and consider all the hours to have the same trend. The issues with the fixed threshold are false and miss detection. This paper proposed hourly basis KPI anomaly detection using machine learning techniques such as supervised learning and outlier detection and to measure the performance of a specific hour because of traffic profile and user behavior differences. The dataset was collected from mobile radio network and the ground truth was determined and labeled by network performance expert. There were 6 selected KPIs with 12096 total data samples including data for training, validation, and testing. 13 machine learning algorithms, 1 statistical technique and 7 data scalers were evaluated. The best performing technique is extra tree algorithm using standard and quantile transformer scaler. With extra tree algorithm, there were 4 missing detections and 7 false detections from 2418 total samples from data testing resulting impressive 97.11% of average F1 score from all 6 KPIs and 3 KPIs are having 100% F1 score. The evaluation result is proof that extra tree algorithm is very suitable for anomaly detection on mobile network hourly basis KPI data and it can significantly reduce the false and miss detection, alongside some general notion of which algorithm is suited for a certain type of KPIs.
Keywords:	Anomaly Detection, Machine Learning, Extra Tree Algorithm, Key Performance Indicator, Mobile Network
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	AN OPTIMAL MACHINE LEARNING MODEL BASED ON SELECTIVE REINFORCED MARKOV DECISION TO PREDICT WEB BROWSING PATTERNS
Author:	Dr. V. V. R. MAHESWARA RAO, N. SILPA, Dr. GADIRAJU MAHESH, SHIVA SHANKAR REDDY, Dr. VIJAYA KUMAR, Dr. K. B. V. BRAHMA RAO
Abstract:	The abundance of user usage data has gained exponential dimensions as a result of the ongoing expansion and spread of Web applications and Web-based systems. Web user usage extraction is used to analyze the browsing data and investigate the web user's visiting interests or patterns. To enhance operational performance, web miners must employ predictive machine learning techniques integrated with reinforced Markov decision process. Especially, Higher order Markov decision frameworks promise the stronger predictive performance and penetration than single-order Markov decision, but they have a large state computational complexity. As a result, a selective Markov decision framework is formulated to considerably increase operating efficiency and prediction accuracy. Towards that the researchers introduced An Optimal Machine Learning Model Integrating with Selective Reinforced Markov Decision process (MLSRM). To efficiently collect and store web browsing data, the MLSRM makes use of the distributed HDFS-Spark parallel computing architecture. It then goes through the necessary pre-processing procedures to get the data ready for the Markov decision process. Later, MLSRM developed a reinforcement strategy to derive actionable knowledge so as to understand online user browsing habits with reduced state complexity and improved forecasting performance by intelligently selecting and integrating several Markov decision processes. Despite of compromising the overall accuracy and proposed model integrity, the suggested methodology eliminates lower Markov support states, examines the awarded probability, and quantifies the error at each Markov state. The proposed prediction Markov decision process was put to several tests, and the results are reported in this article.
Keywords:	Machine Learning; Reinforce Learning; Markov Models; Web Mining; Web User Behavior; Spark; Hadoop; Prediction
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	AN XHAUSTIVE ANALYSIS OF RECENT DEVELOPMENTS IN COMPUTATIONAL EPIGENETICS
Author:	SUKUMAR T, BABY DEEPA V
Abstract:	Objective: To analyze the available data models, recent advancements, and contemporary methodologies in computational epigenetics. Methods: The computational epigenetic modelling of dynamic epigntics was analysed and its efficacy and scope were adjudged. We consulted and compared the available primary graph data sources such as Methy-LogiX, StatEpigen, MethyCancer, PubMeth, MethDB, MethBank, etc. The schema design of Neo4j was analyzed in depth. Specific tools and instruments used in different spectrums of epigenetics were consulted and evaluated. Findings: It has been demonstrated through the use of combinatorial and multivariable analysis that there are design-level barriers in the database structures that prevent the transfer of transcriptional regulatory knowledge from organisms to their cell level. The authors predict that as large databases grow rapidly, current bioinformatic techniques will no longer be sufficient and that fundamental research will be required to develop a new paradigm utilizing high level prototyping and cutting edge bioinformatic techniques. A comprehensive analysis would also be required to determine the likelihood of curing diseases in novel ways using a variety of data sources. Novelty: There are fundamental design level barriers in database structures, and with rapidly growing databases, current bioinformatic techniques are not sufficient. Fundamental research is required to redesign the schema of epigenetic databases and to develop specialized bioinformatic techniques to make the most of the abundance of epigenetic data.
Keywords:	Computational Epigenetics, Dynamics of Epigenetics, Graph Databases, Epigenetic Databases, Epigenomic Mapping.
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	CREDIT CUSTOMER SEGMENTATION WITH HIERARCHICAL CLUSTERING AT VARIOUS DISTANCES
Author:	HENIDA RATNA AYU PUTRI, ADJI ACHMAD RINALDO FERNANDES, ATIEK IRIANY, NURJANNAH4, SOLIMUN
Abstract:	Cluster analysis is a multivariate technique that aims to group objects based on similar characteristics. The purpose of this study was to see the grouping and comparison of the results of cluster analysis using the ward linkage method with various distances (Euclidean, Manhattan and Mahalanobis) with an assessment of the 5C variable (Character, Capacity, Capital, Collateral, Condition of Economy) on Bank X mortgage customers. In this study, we will compare the results of grouping using the three methods based on the ratio of standard deviations between clusters and within clusters. The sampling technique used nonprobability sampling method with purposive sampling basis. The sample size in this study was 100 customers. The standard deviation ratio between clusters and within clusters shows that the results of the Ward method grouping with Mahalanobis distance are better than the results of grouping using other methods. The results of cluster analysis using the Ward method with Mahalanobis distance are cluster one consisting of 26 customers with compliance as the highest aspect, cluster two consisting of 30 customers with condition as the highest aspect, while cluster three consists of 44 customers with capital as the highest aspect. The originality of this study is to compare the distances in hierarchical cluster analysis, especially with ward linkage in segmentation of credit customer.
Keywords:	Cluster analysis, Euclidean distance, Manhattan distance, Mahalanobis distance, Ward's Method
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	KNN-WT BASED COVID-19 DETECTION USING CHEST X-RAY BINARY CLASSIFICATION
Author:	MADHU JAIN, RENU SHARMA
Abstract:	A novel virus commences in Wuhan China in December 2019. It was named as novel coronavirus (nCovid-19) or severe acute respiratory syndrome corona virus-2 (SARS-CoV-2). Due to its zoonotic nature, it had affected animals as well as human beings. The stated virus is spreading at such a rapid rate that it has razed human lives and the global economy. To aid in such pandemic situation, we have proposed a novel neural network-based model for diagnosing coronavirus from a raw chest X-Ray image. The proposed model uses K-Nearest Neighbor (KNN) for classifying the input image. It will support binary classification i.e., COVID effected X-Ray and normal X-Ray. Several collected input images are initially pre-processed using dual-tree complex wavelet transform (DTCWT). Then, feature extraction is executed using mobilenet architecture. Further, image classification is performed using the KNN based model. Lastly, the output is predicted whether it belongs to the Covid-19 class or normal class. For visualizing the effectiveness of the proposed KNN based classifier, parameters such as accuracy, recall, precision, and F1 score are calculated. A comparison is made by calculating the average of all the parameters with existing techniques. Experimental results showed that the proposed KNN-WT model achieves an accuracy of 99%. It outperformed all the existing algorithms.
Keywords:	KNN classifier, DTCWT, chest X-Ray, Image Classification, COVID-19
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	MAСHINE LEARNING TECHNIQUE IN QOS MANAGEMENT NETWORK
Author:	ZHUNUSSOV A., BAIKENOV A., ZHELTAYEV T., SERIKOV T., ZIYEKENOV T.
Abstract:	Modern telecommunication systems and data transmission networks generate large volumes of heterogeneous traffic. In such networks, traditional network management methods for monitoring and analyzing data have some problems in terms of accuracy and efficient processing of big data in real time. The purpose of the study is to show how the value of the number of PPPoE PADT session break packets sent by the router indicates the presence of packet loss along the way. The PPPoE protocol works on the client-server principle, which is mainly used in broadband networks, by encapsulating PPP frames within an Ethernet frame. Using the PPPoE protocol, operators can control the access and operation of subscriber connections, as well as manage a large number of connections at one point in the network. It is proposed to use a machine learning method that will automate the process of tracking the quality of services provided using information from PPPoE sessions in a specific VLAN at specified time intervals. The article analyzes statistics from PPPoE sessions collected in 8 directions of VLANs providing various types of services. Found and fixed problems in the direction of VLAN112, which were associated with signal distortion in the section of the radio relay link along the route. Based on the test results, the proposed method makes it possible to identify a malfunction without involving additional technological resources. To store statistics, you need to allocate memory on the server for each direction and configure PPPoE.
Keywords:	QoS management, PPPoE, PADT, VLAN, Maсhine learning, Monitoring
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	INCREASE RELIABILITY OF PEGASIS MESH NETWORK USING SUBSTITUTION METHOD
Author:	SANTOSH KUMAR YADAV, SANJEEV GANGWAR
Abstract:	Energy improvement and reliable communication are two of the main considerations for building a new routing protocol. Almost every routing protocol uses energy in its data gathering process, chain head selection process, and sensor node failure. To address this sort of issue, we built the PEGASIS mesh replacement approach. In the IRPSM sensor, nodes are grouped into rows and columns, and connections follow the network architecture in the PEGASIS routing algorithms. All nodes have the same capacity, so you can simply pick the master node that is nearest to the base station. The data collecting method is determined by the network topology. The data travels in two steps: the data of the first step is transferred via its own chain to its own head, and the data of the second step is communicated to the nearby node. The first incoming data is accepted, the other data is rejected. Our suggested routing algorithms employ a fallback mechanism in terms of chain head selection. If the top node fails for whatever reason, then the second node becomes the master node. Leveraging the replacement approach and using the network architecture to route PEGASIS, we discover various benefits, such as failure tolerance and overhead concerns. The instantaneous average throughput of the PEGASIS network architecture is marginally better than the chain routing protocol.
Keywords:	Chain Based routing protocol, wireless sensor network (WSN), substitution method, Wireless mesh network, PEGASIS.
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	EVALUATION OF DIFFERENT ATTACKS TOWARD LORAWAN AND SECURITY SUGGESTIONS
Author:	AHMED AL-HTHLOOL, MOUNIR FRIKHA
Abstract:	Nowadays, emerging trends in the field of technology related to big data, cognitive computing, and the Internet of Things (IoT) have become closely related to people's lives. The Internet of Things networks consist of a huge number of interconnected devices and sensors that process and transmit data. Such Activities require efficient energy to be performed at the highest quality and range, hence the concept of Long-Range Wide Area Network (LoRaWAN) introduced, which concerns about delivering lower energy consumption, supporting large networks and mobility. LoRaWAN is a protocol that is designed to connect to operated things to the internet in regional, global networks and target internet of things. LoRaWAN can help make the life of people easier, but it has problems regarding the security such as replay attack, bit-flipping, and others. In this article, the security mechanisms in LoRaWAN will be evaluated as well as simulating security attacks that aims to know how to improve security in LoRaWAN. We discuss security features, cryptographic, key management, message acknowledgement and activation methods of LoRaWAN. Simulation is done under virtual environment using many tools and most used is Mbed simulator. We did the simulation, and the result is the attacks breached to LoRaWAN and got to the network server. Finally, we highlight the mitigation of these attacks.
Keywords:	LoRaWAN, LoRa, cybersecurity, IoT
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	ZERO ORDER AND CONJUGATE IMAGES ELIMINATION FOR DIGITAL HOLOGRAMS
Author:	ZERO ORDER AND CONJUGATE IMAGES ELIMINATION FOR DIGITAL HOLOGRAMS
Abstract:	The zero order and the twin image severely affect the quality of reconstructions in digital holography, as a result of limited resolution and dynamic range of camera sensors. Several techniques have been proposed to minimize or to eliminate those elements. Some are purely numerical; others require experimental interventions. Amongst them are: subtraction digital holography that consists of subtracting a randomly modified recording from the original one, HRO and phase shifting methods that use controlled images for subtraction. We have investigated these methods theoretically and experimentally for both off-axis Fourier and Fresnel holography. Their performance has been assessed with a sample that is a vibrating membrane in the former case, and a stick-shaped transmission diffuser in the latter case. Experimental evidence indicates that if they are efficient for zero-order elimination on off-axis Fourier holograms, a simple digital procedure applied to a single Fresnel recording, and that combines mean value subtraction and spatial filtering in the frequency space permits remove the zero order and the twin image simultaneously.
Keywords:	Quasi-Fourier Off-axis Digital Holography; Fresnel Digital Holography; Filtering Technique; Zero order and Conjugate Image.
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	A NEW AND IMPROVED APPROACH FOR AIR TRAFFIC FLOW PROBLEM IN THE TERMINAL AREA
Author:	EL GHAZI RAOUAN, BENAMEUR LAMIA, CHENTOUFI ALAMI JIHANE
Abstract:	Air transport has always helped humanity and plays a major role in the globalized economic development that the world is currently experiencing; these solutions are practical and cooperative that have erased borders and brought cultures together. Despite its benefits, the increase in air traffic has caused air traffic management problems which has become one of the biggest problems we face and has emerged as an extremely complex problem. This problem involves multiple actors coming to constrain in a complex way. This study proposes innovative solutions to the modeling of conflict resolution and deals with the aircraft scheduling problem at the terminal. as is the NP-hard problem, several heuristic-based approaches are proposed. A new and improved approach Based on Grey Wolf Optimizer and Fireworks Algorithm named FWGWO is applied to solve the air traffic flow problem in the terminal area.
Keywords:	Air Traffic Management, Arrival Scheduling And Sequencing, Fireworks Algorithm, Grey Wolf Optimizer, Optimization Problems, Problem Of Sequencing Aircraft, Metaheuristic
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	A SYSTEMATIC LITERATURE REVIEW: THE POWER OF THE BLOCKCHAIN TECHNOLOGY TO IMPROVE PHARMACEUTICAL SUPPLY CHAIN
Author:	MOHAMED LAHJOUJI, JAMILA EL ALAMI, MUSTAPHA HLYAL, OMAR LAHJOUJI
Abstract:	Despite the blockchain's considerable potential to solve traditional supply chain problems, research on its deployment in pharmaceutical supply chains (PSC) is sparse. Therefore, the objective of this paper is to provide a conceptual framework for blockchain implementation within the pharmaceutical supply chain. To document the twelve-year researchs, 78 transdisciplinary publications published between 2010 and 2022 were examined using a comprehensive literature review and text mining method. Descriptive and thematic research highlights emerging Blockchain trends in pharmaceutical supply chain. Future research will primarily focus on the use of Blockchain for drug counterfeiting and recall issues, as well as other sector-specific challenges like patient health data sharing, compliance, and clinical trials. The arguments and obstacles for technology acceptance, implementation steps and applications highlighted through the thematic analysis will help build the orientation for the research. Compared to other industries, research on blockchain for PSC has lagged, but it has picked up speed since the Covid-19 pandemic. Researchers and professionals will be guided by the identified influencing factors and implementation roadmap for adopting Blockchain in the pharmaceutical business. The suggested framework is original and offers manufacturers, ministry of health, and private sectors helpful guidelines to Leverage the power of blockchain technology.
Keywords:	Blockchain Technology, Pharmaceutical Business, Supply Chain Management, Adoption, Implementation, Systematic Literature Review, Framework, Counterfeiting Drug
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	DISTANCE-BASED COMPARISON OF HOPFIELD NEURAL NETWORK SEGMENTATION USED IN LIVER REGION EXTRACTION FROM CHEST CT IMAGES
Author:	KHAWLAH ALBELIHI, RACHID SAMMOUDA, ABDULMALIK ALSALMAN
Abstract:	Liver cancer is a silent disease because most patients do not have symptoms or indications in the early stages. Early diagnosis is necessary to detect the disease and hence prevent its progression. The aim of our computer-aided detection (CAD) system is to detect liver cancer. In this paper, we study the effect of the distance measure type on the segmentation results of the Hopfield Artificial Neural Network (HNN), which is used to extract the liver region from chest Computed Tomography (CT) images without any pre-processing. We compare three distance measures: the Euclidian distance measure, the Standard Euclidian distance measure, and the Manhattan distance measure. The Euclidian distance measure shows the best segmentation result for the liver region. It also has the best performance in terms of its energy minimization function.
Keywords:	Hopfield Neural Network, Artificial Intelligence, Liver Cancer, CAD, Segmentation, Distance Measure.
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	RECOMMENDATION SYSTEM FOR ONLINE JOB VACANCY USING MACHINE LEARNING MODELS
Author:	YOSUA F SIMANJUNTAK, ANTONI WIBOWO
Abstract:	The Recommendation is used to exploit the relations among known features and content that describe items (content-based filtering) or the overlap of similar users who interacted with or rated the target item (collaborative filtering). To combine these two filtering approaches, current model-based hybrid recommendation systems typically require extensive feature engineering to construct a user profile. Name Entity Recognition (NER) provides a straightforward way to identifies one item from a set of other items that have similar attributes of the related objects. However, due to the large scale of the data used in real world recommendation systems, little research exists on applying NER models to hybrid recommendation systems in job vacancy environment. This paper is proposed a way to adapt the name entity recognition approaches to construct a real hybrid job recommendation system. Furthermore, in order to satisfy a common requirement in recommendation systems the approach of accuracy, precision, recall and F-measure is using in this recommendation system in a principled way. The experimental results demonstrate the efficiency of our proposed approach as well as its improved performance on recommendation precision.
Keywords:	Recommendation System, Job Recommendation, Content-based filtering, Name Entity Recognition, TF-IDF
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	A QoS BASED PROGRESSIVE LINK PREDICTION MATHEMATICAL APPROACH FOR DATA TRANSFER IN MANET
Author:	A.MAHENDRAN, Dr.C.KAVITHA, K.SAKTHIVEL
Abstract:	The MANET (mobile ad hoc network) is a self-designed, self-coordinated, foundationless network comprising mobile nodes. Every node can move the data packets with one another over one or the other radio or infrared. All networks should give an acceptable and helpful degree of Quality of Service (QoS) to guarantee that applications are very much upheld. This becomes a challenge regarding Mobile ad-hoc networks (MANETs). To help the developing requirement for multimedia and constant applications, quality of service (QoS) support by the networking protocol is required. A few significant QoS boundaries that are required by such applications can be distinguished. They incorporate transfer speed, start to finish delay, delay jitter, and spot blunder rate. So in this paper, we proposed a QoS based Progressive Link Prediction Mathematical Approach for Data Transfer in MANET. Model to estimate the future status of link availability using a QoS based Progressive Link Prediction Mathematical Approach for Data Transfer in MANET. A mathematical model for preventing Attacks in MANET by Using Elliptic Curve Cryptography Algorithm. The experimental result based on the proposed model with AODV routing is better than the existing methods.The proposed model combines a mathematical approach and ECC, along with AODV routing, for predicting links and preventing attacks. The main objective of this research is to provide an efficient data transfer using an optimized prediction and attack prevention model.
Keywords:	MANET, QoS, Link prediction, data transfer, mathematical approach, AODV routing.
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	5G TECHNOLOGY AND ITS IMPACT ON THE USE OF ONLINE VIDEOGAMES: A COMPREHENSIVE SYSTEMATIC REVIEW
Author:	JAVIER GAMBOA CRUZADO, ANTHONY PAYI QUISPE, CRISTINA ALZAMORA RIVERO, AMANDA DURAN CARHUAMACA, MARIA RODRIGUEZ KONG, JORGE NOLASCO VALENZUELA
Abstract:	5G technology has received a lot of attention lately thanks to increased connections, synchronization between devices, better speed and higher bandwidth. But it still presents challenges, such as infrastructure and data security. On the other hand, we live in an era where being connected to the Internet seems intrinsic in our lives, the future says that "the Internet of things" will be present even in devices that we would never have imagined, and in the use of video games there are fewer and fewer titles that do not make use of the network in one way or another. This research aims to determine the state of the art on recent studies and provides an overview that relates 5G technology and Online Videogames, and will help to close existing gaps and direct future research. The Systematic Literature Review (SLR) located 9032 papers from major digital libraries such as: Scopus, Wiley Online Library, ARDI, Taylor & Francis Online, ERIC, ProQuest, ACM, Google Scholar and Springer. The research has allowed us to identify clusters of papers whose Conclusions and Discussions are Objective, Bibliometric Networks of Keywords, named entities (NER) that are most frequently presented in the Abstracts and the most cited topics in the experimental research reviewed on the subject. After a detailed review process, only 70 papers were considered according to the defined inclusion and exclusion criteria.
Keywords:	Game, Online video games, 5G technology, Systematic Literature Review
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text

Title:	SYSTEMATIC LITERATURE REVIEW OF DECISION SUPPORT SYSTEM FOR SOCIAL MEDIA
Author:	MUHAMMAD FIKRI HASANI, ERNA FRANSISCA ANGELA SIHOTANG, GALIH DEA PRATAMA, AFDHAL KURNIAWAN, DITDIT NUGERAHA UTAMA
Abstract:	Social media is a very active and fast-moving domain. It is not only a communication medium but also for information exchange. Based on all categories of sentiment interchanged in media, the central issue could be academically addressed and can be methodically converted to decision support. The study aimed to review the decision support system on social media, including structural issues and the data techniques to extract the estimated valuable information. This research used Google Scholar, IEEE, Elsevier, Research Gate, and Semantic Scholar online databases for articles published in English during the last nine years (between 2012 and 2021). We specifically searched for three keywords ("Social Media", "Decision Support" and "Decision Support of Social Media") to find the articles. In this literature, we presented a systematic review of the decision support system using three research questions. The final result is a continuity between a system of social media until being a system for decision support. So many fields support decisions based on social media facts and databases. The decision could support the decision unit for determining a big decision on their business or organization. Many scientific methods would make the decision, and varied methods and cases took the case selection. This research is expected to contribute to academic research on social media decision support.
Keywords:	Decision Support System, Literature Review, Social Media
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2023 -- Vol. 101. No. 2-- 2023
Full Text