		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
March 2021 | Vol. 99 No.06

Title:	RATIONALIZING THE CHOICE OF ACADEMIC MAJOR FOR UNDERGRADUATES USING CLASSIFICATION ALGORITHMS
Author:	ELMASRY, MOHAMED ABBAS
Abstract:	Lately, there has been a massive increase in the amount of data stored in educational databases. These databases include important knowledge of student score records. Educational data mining is used to study the data available in the educational field and extract unseen implications from it. From the moment a student enrolls in higher education, he needs guidance and assistance in decision making. Choosing a major is one of the important decisions for students to be qualified for. The decision to choose a major should be based on a rationale suitable for the student's cognitive skills. It should be noted that choosing a major is also important because it is often related to the type of job the student will occupy after graduation. All of these considerations require detailed research so that we can scientifically guide students, especially the most vulnerable students, to encourage them to realize their fullest potential in the major that best suits them. So, to what extent can the current academic advising system help students make this crucial decision? Academic advisors need a clear form that sheds light on how to guide students in choosing the right major for them. Looking at the rate of change in majors for students of the faculty of Business Administration of the National Egyptian E-Learning University (EELU), it was found to be a very high rate during the six-year period covered by the study, reaching 48% of the students, which reflects the lack of clarity of vision for these students while they make a decision about the major. Classifying student data would contribute to making decisions about making rational decisions regarding major selection, and achieve the best results. In this paper, we used classification algorithms for a dataset of business administration students at the National Egyptian E-Learning University in a way that enables academics to predict the student’s GPA in different majors, so that students can use them to make informed decisions about the appropriate major choice. The model used in this study predicted the appropriate major with an average recall value of 89.83%. Faculty members can benefit from these results in taking the necessary actions to improve the student's academic performance. This study may reduce dropout rates and improve student performance in the recommended majors.
Keywords:	Educational Data Mining, EDM, Classification Algorithms, Prediction, Major Selection, Undergraduates, Higher Education, Academic Performance.
Source:	Journal of Theoretical and Applied Information Technology 31^st March 2021 -- Vol. 99. No. 06 -- 2021
Full Text

Title:	ENTERPRISE ARCHITECTURE MEASUREMENT: A SYSTEMATIC LITERATURE REVIEW
Author:	AMMAR ABDALLAH , ALAIN ABRAN , MOHAMMED A. KHASAWNEH
Abstract:	A number of measurement solutions have been proposed to manage the development of Enterprise Architectures (EA) but this body of knowledge has not been analyzed to identify any strengths or weaknesses from a measurement perspective. Adopting a systematic literature review (SLR) approach this research identified 23 primary studies on EA measurement solutions. These studies were analyzed through five research questions looking into the types of EA entities being measured, sources and types of input data, mathematical operations, and measurement units used. The key findings are: 1) the selected studies measure four entity types - EA as an architecture, as a framework, as a project and as a program; 2) most of the input data to the measurement solutions are subjective data reflecting practitioner opinion; 3) mathematical operations used in the measurement solutions are mostly scores and weight multiplications without verification for valid mathematical operations; 4) most often they lack well-defined measurement units.
Keywords:	Enterprise Architecture (EA), Software Engineering Measurement, Metrology, Metrics.
Source:	Journal of Theoretical and Applied Information Technology 31^st March 2021 -- Vol. 99. No. 06 -- 2021
Full Text

Title:	KNOWLEDGE MANAGEMENT CATALYSTS IN TERTIARY INSTITUTIONS
Author:	A. S. OGUNBANWO, J. O OKESOLA, AYOADE A. ADEBIYI, SHERYL BUCKLEY
Abstract:	Knowledge Management (KM) has become a prerequisite to institutional performance to the extent that its adoption is now a tool for institutional competition amongst tertiary institutions in Nigeria. However, it has failed to achieve organizational objectives in some instances because underlining contributory elements to effective KM implementation are not being adequately considered. Although many factors have been traced to this failure, most of them are yet to be empirically tested and are therefore not generally acceptable. This study recognizes KM enabling capability and KM strategy capability as success factors to KM, and adopt a survey over 10 Nigerian universities to empirically test their contributions. SPSS software was employed for demographic data analysis over frequency count and percentage score, while the hypotheses were analyzed with Pearson correlation and multiple linear regression. The results show a positive linear relationship between the elements of KM enabler, KM strategy and KM success confirming KM enablers and KM strategy as positive contributors and catalysts to KM success in Nigeria South West tertiary institutions.
Keywords:	Enabler capability, Knowledge management, KM Success, Process capability, Strategy capability
Source:	Journal of Theoretical and Applied Information Technology 31^st March 2021 -- Vol. 99. No. 06 -- 2021
Full Text

Title:	A NOVEL APPROACH OF CONGESTION MANAGEMENT IN TRANSMISSION NETWORKS USING AN ADVANCED INTERLINE POWER FLOW CONTROLLER WITH CONSTRICTION FACTOR-BASED PARTICLE SWARM OPTIMIZATION ALGORITHM
Author:	B. BADDU NAIK, CH. PADMANABHA RAJU, R. SRINIVASA RAO
Abstract:	This paper presents a viable strategy for congestion management in power frameworks. Congestion in the power transmission networks is one of the specialized issues that show up in the electric power framework. Power dispatch is one of the significant control exercises and the Optimal Power Flow (OPF) is the main apparatus to acquire the least expense generation designs with transmission and operational requirements. Clog is eased utilizing without price techniques. Utilizing Flexible AC Transmission (FACTS) gadgets, blockage can be decreased without upsetting the financial issues. Advanced model of Interline Power stream Controller (AIPFC) is predominantly an arising FACTS gadget and it is utilized in this paper to decrease the clog. In clog management, the target work is nonlinear subsequently in settling this capacity a novel Constriction Factor-Based Particle Swarm Optimization (CFBPSO) algorithm was proposed for blockage the executives with the point of expanding social government assistance while minimization of generation price. CFBPSO strategy with AIPFC is tried on IEEE 30-bus framework and it tends to be stretched out to any down to practical framework. The outcome for the quality framework was gained by reproducing the check framework utilizing MATLAB/SIMULINK.
Keywords:	Flexible AC Transmission System (FACTS), Advanced Interline Power Flow Controller (AIPFC), Optimal Power Flow (OPF), Constriction Factor Based Particle Swarm Optimization (CFBPSO).
Source:	Journal of Theoretical and Applied Information Technology 31^st March 2021 -- Vol. 99. No. 06 -- 2021
Full Text

Title:	3D MODELLING BY MEANS OF ARTIFICIAL INTELLIGENCE
Author:	BEBESHKO B., KHOROLSKA K., KOTENKO N., DESIATKO A., SAUANOVA K., SAGYNDYKOVA S., TYSHCHENKO D.
Abstract:	For years humanity has developed some solid perception of the world around them. According to this perception, it is easy to describe complex structures through a short explanation. For example, by telling a person to “think of a yellow submarine, flying in the sky”, the person in question would have an exact image of the machine you are talking about, without having seen it ever before. Along with this - artificial intelligence becomes a rapidly growing area of research, and it is possible that artificial intelligence will make it possible for computers to gain perception of the world. Artificial intelligence can be used to solve problems without the need to specify how to solve the settled task. This paper outlines features of 2D images recognition and the creation of 3D models, using AI and machine learning accordingly. Therefore, it would make designers work much easier, - graphic designers and architects would be able to only specific features instead of spending a lot of time on the actual drawing. Moreover, if models were computationally generated, the developing process would become much more simple and could result in a higher developer’s performance rate. The key purpose of this work is to define the possible limitations of CNN usage for 3D models generation, taking into account output resolution and generation swiftness.
Keywords:	Artificial Intelligence, Neural Network, 3D, 2D, Image, Models, Graphics
Source:	Journal of Theoretical and Applied Information Technology 31^st March 2021 -- Vol. 99. No. 06 -- 2021
Full Text

Title:	A COMPARATIVE ANALYSIS PERFORMANCE OF DATA AUGMENTATION ON AGE-INVARIANT FACE RECOGNITION USING PRETRAINED RESIDUAL NEURAL NETWORK
Author:	KENNEDY OKOKPUJIE, ABERE REUBEN, JOYCE C. OFOCHE, BASUO J. BIOBELEMOYE, IMHADE PRINCESS OKOKPUJIE
Abstract:	There has been an immense improvement in face recognition research. Unfortunately, the accuracy of face recognition systems recognizing the same person over time due to ageing is open research. Minor geomet-ric changes in the face that occur due to ageing contribute to face recognition systems' inaccuracy. Re-searchers, over subsequent years, have come up with methods to improve the performance of Age Invariant Face Recognition (AIFR) systems, the most recent one being the use of Convolutional Neural Network (CNN) to create face recognition models. The pre-trained residual network (ResNet) is trained and tested using a heterogeneous database to actualize this improvement. The heterogeneous database consists of images from 82 Caucasian subjects in the FG-Net database and 11 African subjects. These obtained images were augmented using geometric transformation and Noise to increase the amount of data for training. Af-terwards, a model robust is developed. The Sliding Window framework was used to detect the faces fed into CNN for training and testing. After getting the results from our classification model, an analysis was carried out on the classification models of both the original dataset and augmented datasets. It was ob-served that the model performed remarkably with the noise-injected dataset and performed worst with the geometric transformation database.
Keywords:	Age Invariant Face Recognition (AIFR), Convolutional Neural Network (CNN); Geometric Transformation; Noise injection; Data Augmentation
Source:	Journal of Theoretical and Applied Information Technology 31^st March 2021 -- Vol. 99. No. 06 -- 2021
Full Text

Title:	IMPROVE THE PERFORMANCE OF TRANSPORT LAYER OF TCP PROTOCOL IN WIRELESS SENSOR NETWORK
Author:	SAIF TALIB ABBAS ALBASRAWI
Abstract:	The Wireless Sensor Network is a network of many sensor nodes with wireless channels to communicate with others. With no centralized control and specified network connection, it can transfer to the outside world. At the same time, both nodes are capable of acting as a source or sink node. These nodes have minimal processing power due to their small physical size, limiting the processor capacity and the battery's size. When they work together collectively, they can gather information about the physical environment. They get a transceiver to communicate with the virtual world and with the real world. The routing topology to be used for the network depends on the transmission capacity of its nodes. It also depends on the location of the node, which may differ from time to time. This paper examines the characteristics of a wireless sensor network and analyzing the functionality of TCP protocol and its variables. It suggests using nodes near the sinks as proxy nodes to improve the WSN transport layer's efficiency. The results of the simulation show that the throughput increases.
Keywords:	TCP protocol, WSN, AODV, Network Simulator NS-2
Source:	Journal of Theoretical and Applied Information Technology 31^st March 2021 -- Vol. 99. No. 06 -- 2021
Full Text

Title:	AEGD: ARABIC ESSAY GRADING DATASET FOR MACHINE LEARNING
Author:	BASSAM AL-SHARGABI, RAWAN ALZYADAT, AND FADI HAMAD
Abstract:	Recently, developing an Automatic Essays Grading (AEG) system has become an attractive topic in industry and academia. Most of the grading systems rely on machine learning to grade the essays based on a predetermined dataset. However, English essays scored based on Automated Student Assessment Prize (ASAP) dataset whereas the absence of such a dataset for Arabic essays is a major predicament. Therefore, in this paper, we have established the Arabic Essay Grading Dataset (AEGD) that is suitable for machine learning to develop an Arabic AEG system. This dataset comprises a collection of essay questions along with its graded model answers for several topics that cover various school levels. We used the Naive Bayes (NB), Decision tree (J48), and meta classifier as a well-known machine learning algorithms to evaluate and test the established AEGD. The results show that the accuracy rates of the three classifiers have reached 79%, 81%, and 86% based on the established AEGD..
Keywords:	Automated Essay Grading; Arabic Essay Grading; Dataset; Machine Learning; Classification Algorithm
Source:	Journal of Theoretical and Applied Information Technology 31^st March 2021 -- Vol. 99. No. 06 -- 2021
Full Text

Title:	DETECTING COMMON WEB ATTACKS BASED ON SUPERVISED MACHINE LEARNING USING WEB LOGS
Author:	XUAN DAU HOANG, TRONG HUNG NGUYEN
Abstract:	Web attacks, such as SQLi (SQL injection) and XSS (Cross Site Scripting) have been seen critical threats to web applications, websites and web users. These types of web attacks can cause serious damages to web applications, websites and web users, ranging from bypassing authentication systems, stealing sensitive information from databases and users, to even taking the full control of server systems. To cope with web attacks, a number of methods have been researched and applied to protect web applications, websites and web users. Among them, the detection of web attacks is a promising approach in defensive layers to safeguard websites and web applications. However, some methods can only detect one kind of web attacks, while other proposals either require regular updates of detection rules, or require extensive computing resources because they use complicated detection methods. In this paper, we propose a model for web attack detection based on machine learning using web logs. Our model’s main aims are (1) building the detection model automatically and without the requirement of frequent update, (2) being able to detect common types of web attacks and (3) improving the detection rate as well as lowering down the false alarm rates. The proposed detection model is built using inexpensive machine learning algorithms, including SVM, decision tree and random forest. Experiments conducted on a labelled dataset and real web logs show that the proposed model is capable of detecting common types of web attacks effectively with the highest overall detection accuracy rate of 99.68%.
Keywords:	Common Web Attacks, Web Attack Detection, SQL injection Detection, Cross Site Scripting Detection, Machine Learning-based Web Attack Detection
Source:	Journal of Theoretical and Applied Information Technology 31^st March 2021 -- Vol. 99. No. 06 -- 2021
Full Text

Title:	THE INFLUENCE OF COVID-19 ON E-COMMERCE TOWARDS ONLINE SHOPPING
Author:	AHMED I. EL SEDDAWY, MOHAMED HEGAZY MOHAMED
Abstract:	The COVID-19 emergency is resuscitating a growth of online business towards new firms, clients and sorts of things, likely including a drawn out move of web business exchanges from luxuriousness item and endeavors to standard necessities. It in addition incorporates how strategy producers can use the limit of front line change in retail and related regions to help business assortment and to update social secluding, while at the same time guaranteeing that nobody is abandoned. Retail and food association’s deals among February and April 2020 were down 7.7% stood apart from a near period in 2019. In any case, deals stretched out for business areas and non-store retailers (all things considered online business providers) by 16% and 14.8% freely. Subsequent to running the model the exhibition demonstrate that the precision of the CHAID model is 89.09% and Classification Error is 10.91% this is the best Operator for anticipating Types of Goods that purchasers will purchases as contrasted and Decision Tree and Random Tree as showing in Table 8 and showed in Figure 11. The results of this contextual analysis plainly demonstrate that CHAID is appropriate porter for identifying Types of Goods that customer demeanor for taking choice for buying or Not. One preliminary for predicting buyer lead for E-exchange online through 1000000 models and 8 apparent credits. The expert parceled the data to rule regions at first is getting ready data equal 90% and second is attempting data equal 10% In the wake of running the CHAID pattern, the CHAID made as appeared in Figure 4 by Rapid miner Tool for Invoice Types for items is the most Attributes in all Attributes. Execution vector CHAID Operator for Types of Goods quality showed in Figure 6. Precision CHAID Model showed in Figure 4 Classification mistake CHAID Model showed in Figure 5. Disarray Matrix CHAID Model showed in Figure 6. X Plot CHAID Model showed in Table 4. Accuracy Decision Tree Model showed in Table 5. Characterization mistake Decision Tree Model showed in Figure 9. Disarray Matrix Decision Tree Model showed in Figure 8. X Plot Decision Tree Model showed in Figure 8. Accuracy Ranom Tree Model showed in Table 7. Order blunder Ranom Tree Model showed in Figure 9. Disarray Matrix Ranom Tree Model showed in Figure 9. X Plot Ranom Tree Model showed in Figure 10.
Keywords:	Five Covid-19, CHAID Model, Decision Tree Model, Data Mining, E- Commerce
Source:	Journal of Theoretical and Applied Information Technology 31^st March 2021 -- Vol. 99. No. 06 -- 2021
Full Text

Title:	BUSINESS STUDY OF NETWORK PROVIDER DEVELOPMENT IN XYZ INDUSTRY AREA WITH NNI MODELING (NETWORK TO NETWORK INTERFACE) AS A STAGE TOWARDS SMART INDUSTRIAL PARK
Author:	EDI YUSUF WIRAWAN, RIYANTO JAYADI
Abstract:	Network to Network Interconnection business model to build a network provider in an industrial area requires careful calculation and design because it is related to multi-provider networks and, particularly, related to investments and operations that must be profitable in the future.It is adopting Metro Ethernet technology as the backbone and FTTX technology as the access transmission system of the Network to be interconnected. The data analysis using top-down with the business case as the objective and is translated into network design and interconnection modeling between providers from outside the industrial area, and is carried out qualitatively and quantitatively. The discussion starts with the needs of end customers, marketing strategies and is manifested in a network design that can answer all the requirements of tenants in industrial areas and support the acceleration of the smart industrial park.An accurate investment calculations, by selecting the right backbone technology and access can be a successful factor in building network provider in industrial estates.
Keywords:	Network Service Provider, Fiber Optics Networks, Business Modeling, Network To Network Interconnection, Financial Analysis, Service Level Agreement
Source:	Journal of Theoretical and Applied Information Technology 31^st March 2021 -- Vol. 99. No. 06 -- 2021
Full Text

Title:	DEVELOPMENT OF METHODS FOR CHANGING THE RESOLUTION OF IMAGES OBTAINED FROM SMART CITY CCTV CAMERAS BASED ON ALGEBRAIC CHARACTERISTICS
Author:
Abstract:	Article Retracted Due to Plagerism These results have already been published earlier in the works: 1. Single-frame image super-resolution based on singular square matrix operator. ELECTRICAL AND COMPUTER ENGINEERING (UKRCON): proc. of the IEEE First Ukraine conf., May 29 – June 2 2017, Kyiv, Ukraine. – IEEE, 2017. – P.944–948. - https://ieeexplore.ieee.org/document/8100390 2. http://ena.lp.edu.ua:8080/bitstream/ntb/34713/1/27_197-206.pdf 3. Two-frames image superresolution based on the aggregate divergence matrix. Data stream Mining & Processing: proc. of the 1st international scien. and techn. conf., 23–27 August 2016, Lviv, Ukraine. – Lviv: Lviv Polytechnic Publishing House, 2016. – P.235–238. - https://ieeexplore.ieee.org/document/7583548 4. Image superresolution via divergence matrix and automatic detection of crossover, International Journal of Intelligent Systems and Applications (IJISA), Vol.8, No.12, pp. 1-8, 2016. DOI: 10.5815/ijisa.2016.12.01
Keywords:
Source:	Journal of Theoretical and Applied Information Technology 31^st March 2021 -- Vol. 99. No. 06 -- 2021
(Article Retracted ) Full Text

Title:	SENTIMENT-BASED MACHINE LEARNING AND LEXICON-BASED APPROACHES FOR PREDICTING THE SEVERITY OF BUG REPORTS
Author:	ALADDIN BAARAH , AHMAD ALOQAILY, ZAHER SALAH, ESRA’A ALSHDAIFAT
Abstract:	Fixing bug reports is an important activity performed during software maintenance. End-users and software developers report bugs related to open and closed-source projects through a bug tracking system. They should describe the bug reports carefully, mainly when they assign the severity of the bug. Thus, assigning incorrect severity levels will postpone the fixing order of critical bugs. Many works have been proposed using various machine learning algorithms to classify the severity of bug reports. However, few research works have considered the analysis of reporters sentiments expressed in the summary description of bug reports to predict the bug severity. In this paper, the analysis of the reporters sentiments has been considered and incorporated into the severity prediction process. More specifically, sentiment-based approaches have been proposed, namely machine learning and lexicon-based approaches for predicting the severity of bug reports. SentiWordNet dictionary is used to identify the bug reports sentiment terms and compute their associated sentiment scores. The proposed sentiment-based approaches have been applied and evaluated on a bug reports dataset related to closed-source projects extracted from the JIRA bug tracking system. The results of sentiment-based machine learning and lexicon-based approaches are compared and reported. The results have shown that the Logistic Model Trees (LMT) model outperforms other sentiment-based models, including the lexicon-based model. The reported experimental results also revealed that the lexicon-based approach is not effective for bug severity prediction. However, the sentiment-based machine learning approach significantly improves the severity prediction of bug reports compared to the lexicon-based approach (baseline approach). The severity prediction accuracy has been remarkably enhanced from 53% for lexicon-based to 87.14%. Likewise, the F-Measure of the severity prediction has been improved from 0.65 for lexicon-based to 0.91 after applying the machine learning approach.
Keywords:	Software Maintenance; Bug Report; Severity Prediction; Sentiment Analysis; Machine Learning Algorithms; Lexicon-Based; Sentiment-Based Approach.
Source:	Journal of Theoretical and Applied Information Technology 31^st March 2021 -- Vol. 99. No. 06 -- 2021
Full Text

Title:	DEVELOPMENT OF MODELS OF THE MAXIMUM POWER OF THE SOLAR ENERGY TRACKING SYSTEM BASED ON A PHOTOPANEL
Author:	F.SATYBALDIYEVA, R.BEISEMBEKOVA, A.SARIBAYEV, G.Zh. YESSENBEKOVA, A.KULTAS
Abstract:	Solar tracking system is the most common method to improve the efficiency of solar PV modules. This study presents the energy conversion efficiency of a PV module with a solar PV module tracking system. The proposed model of the solar tracking system uses MPT. With the help of the MPT, the solar panel becomes more energy efficient, and with the use of the tracking system it is more sensitive to the sun's rays, and this allows for a more accurate detection of the sun location. A comparative analysis was carried out between a stationary PV module and a PV module with a tracking system. The results showed that a two-axis solar tracking system power production has increased by a total of 1.5%. compared to power produced by stationary PV module. Additionally, the average efficiency of our PV module varies within 12-20%. But still, in the age of rapid development of science and technology, these indicators are not the limit. There are a number of alternative ways to increase these numbers which many scientists from different countries are working hard. In our study, in order to increase the efficiency of the photo module, we considered the optimization of the production of a solar battery with tracking system, which should track the Sun. Because the incidence of the sun rays on the surface of the panel at 90 degrees increases the efficiency of the production of PV modules. The article considers the software and hardware for the control of the solar panel with a rotating mechanism, and solves scientific and applied problems for mathematical modeling and information support of the photo panel (PP) with a solar control system in order to increase PP energy efficiency. The photo panel model was implemented in the Lab View MatLab software package through the Matlab block library.
Keywords:	Photopanel, Support Rotary Mechanism (SRM), Solar Energy, Maximum Power Generation Technology (MPT)
Source:	Journal of Theoretical and Applied Information Technology 31^st March 2021 -- Vol. 99. No. 06 -- 2021
Full Text

Title:	REAL-TIME BIG DATA CLUSTERING USING SPARK: UBER CASE STUDY
Author:	M. EL-SAID EL-BARBEER, AMIRA REZK, A. ABU-ELFETOUH SALEH
Abstract:	We live in a world flooded with data. Some even see it as the fuel that drives all companies to reach their goals. Business Intelligence is introduced to enable a company to get the power of its data to be able for the competition in the rougher market. Because business needs to make decisions in a fast and reliable manner, analysis the big data in real time become interested issue. Although the significant efforts that done in this area, big data analysis in real time is still need additional effort to enhance the performance and reduce the required time. This paper introduces a framework to analyze big data in real-time using the K-means clustering technique. Although the K-mean is widely used in clustering, its processing requirement can be a problem in big data and real-time systems. In this research, the K-mean algorithm is adapted to be suitable for the case of big data and real-time systems. The proposed framework introduces two models the first one uses historical data to create a model which deployed to real-time data and the second one analyzes the data in real-time without historical data. Experimental results show that the accuracy of the proposed framework with its two models is approximately 0.5, 0.34 respectively using the Silhouette Coefficient measurement.
Keywords:	Business Intelligence (BI), Big Data, Real-time Analytics, Clustering, Apache Spark
Source:	Journal of Theoretical and Applied Information Technology 31^st March 2021 -- Vol. 99. No. 06 -- 2021
Full Text

Title:	PREEMPTIVE-RESUME PRIORITY QUEUE SYSTEM WITH ERLANG SERVICE DISTRIBUTION
Author:	O.R AJEWOLE, C.O. MMDUAKOR, E.O ADEYEFA, J.O. OKORO, T.O. OGUNLADE
Abstract:	This paper describes a preemptive-resume priority queue system that assumes Poisson arrivals and a single server facility. Various priority queueing models have been proposed in literature to explain the behaviour of different kinds of queues commonly observed at various service facilities. Majority of these models assume exponential service times. However, when the service time of a given facility is processed in more than one stage and service is in a sequential order (an often encountered scenario in practical situation), the need for a service distribution that can represent this becomes necessary. Hence in this study, the service time distribution is assumed to have Erlang service times. It is assumed that there are two classes of priority levels of which one has preemptive-resume priority over the other. The mean value theorem is applied in determining the performance measures of the higher priority queue. The busy period of the higher priority class assuming First Come First Serve principle and its associated moments is derived. We also evaluate the average number of customers in the system for the lower priority level and other performance measures like the mean sojourn time in the system. Subsequently, the impact and significance of preemptive scheduling is investigated with the use of real life data.
Keywords:	Preemptive-Resume Priority Queue, Erlang Service Distribution, Completion Time, Busy Period, Single Server.
Source:	Journal of Theoretical and Applied Information Technology 31^st March 2021 -- Vol. 99. No. 06 -- 2021
Full Text

Title:	THE DIRECT AND INDIRECT RELATIONSHIPS AMONG A BANK'S WEBSITE DIMENSIONS, PERCEIVED FLOW, PERCEIVED PLAYFULNESS, E-BANKING SATISFACTION AND E-BANKING LOYALTY
Author:	MOHAMED SALIH YOUSIF ALI, KAMAL MOHAMED HAMID ALFAKI
Abstract:	The aim of this research is to investigate the direct and indirect relationships between banks’ website quality (BWSQ) dimensions, perceived flow (PFL), perceived playfulness (PPN), e-banking satisfaction (EBS) and e-banking loyalty (EBL) perceived by Saudi Arabian e-banking service users using the stimulus-organism-response approach. The primary data were gathered by a questionnaire survey using convenience sampling. A total of 336 usable questionnaires were returned. The collected data were analyzed using SPSS 25 and AMOS 25. The results revealed that there are positive relationships between website quality dimensions and PPN, PFL, EBS and EBL. PPN and PFL play a mediating role between BWSQ and EBS. Furthermore, EBL mediates the PPN and EBL relationship. The study closes with a discussion, including implications, limitations, and the direction of future studies, and a conclusion.
Keywords:	Users loyalty, Satisfaction, perceived flow, perceived Playfulness, Website Quality
Source:	Journal of Theoretical and Applied Information Technology 31^st March 2021 -- Vol. 99. No. 06 -- 2021
Full Text

Title:	AN APPROACH FOR COVID-19 DETECTION USING DEEP CONVOLUTIONAL FEATURES ON CHEST X-RAY IMAGES
Author:	ZUHERMAN RUSTAM, SRI HARTINI, ILSYA WIRASATI, JANE EVA AURELIA
Abstract:	First screening of COVID-19 becomes very crucial because of its fast spread. There are several ways to diagnose someone who has COVID-19, but chest X-ray is one of the efficient tools that can be used. Deep learning, especially Convolutional Neural Network (CNN), is commonly utilized in medical images due to its superiority in extracting high-level features of images. However, in order to train CNN, we need enormous data to avoid overfitting. Meanwhile, there is a limit of chest X-ray availability that can be access publicly. Considering this problem, we propose pre-trained CNN model as a feature extractor, and the feature vector obtained as the output of CNN that is used as the input of machine learning classifier, namely Support Vector Machines (SVM), Random Forest (RF), and k-Nearest Neighbors (kNN). Using the data from Kaggle COVID-19 Radiography Database, our proposed method with SVM as a classifier succeeded in delivering accuracy of 99.73% in the testing data. Moreover, the performance of CNN-SVM held on training data provides the average accuracy of 99.77%. Thus, our proposed approach can be used as an alternative on screening COVID-19.
Keywords:	Convolutional Neural Network, Feature Extraction, Hybrid Method, Medical Image
Source:	Journal of Theoretical and Applied Information Technology 31^st March 2021 -- Vol. 99. No. 06 -- 2021
Full Text

Title:	PRELIMINARY INSIGHTS INTO CYBERSLACKING IMPACT ON GRADUATE STUDENTS ACADEMIC PERFORMANCE: A CASE STUDY OF A BUSINESS SCHOOL IN GHANA
Author:	ACHEAMPONG OWUSU, EDITH MAWULI AFI BLEBOO, IVY HAWAH TAANA
Abstract:	The main objective of the study was to examine the impact of cyberslacking on graduate students’ academic performance at a Business School in Ghana. The study was descriptive and was purely quantitative. The target population of this study comprised all graduate students at the School. Out of the entire population, three hundred (300) students were sampled for the study through convenience sampling. Questionnaires were used as the data collection tool. Findings from the analysis indicate that Cyberslacking correlates with students’ academic performance. Students who are addicted to cyberslacking have difficulties in paying attention in class in comparison to those who do not cyberslack. The study, therefore, concludes that though cyberslacking has a negative effect on the attention of graduate students in the lecture room, it could not find a significant relationship between cyberslacking and academic performance. It is however recommended that instructors should integrate technology procedures in their curricula, explain their motivations, and enforce them. Also, the management of universities should ensure graduate students are mindful of their multitasking limits and cyberslacking’s negative effect on learning. Although cyberslacking negatively influences attention and even student learning, several college students underrate this concern because they overrate their capacity to multitask.
Keywords:	Cyberslacking, Graduate Students, Academic Performance, Business School, Ghana
Source:	Journal of Theoretical and Applied Information Technology 31^st March 2021 -- Vol. 99. No. 06 -- 2021
Full Text