		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

June 2023 | Vol. 101 No.11

Title:	IMPLEMENTATION OF SURVEY MOBILE APPLICATION: A VOTING ALGORITHM FOR SOCIAL INFLUENCE MINIMIZATION USING GPS
Author:	ABDULRAHMAN ALKANDARI, NAYEF ALAWADHI, AHMED ALONAIZI, ABDULLAH ALSHEHAB, DALAL ALMUTIRI
Abstract:	Old approaches methods for voting or survey systems are becoming outdated in today's world, and recent improvements are being introduced, the survey applications are getting revolutionized with technological development. The mobile survey applications allow users to cast their votes and provide feedback with an easy and trust-enabled method. Surveys are increasingly becoming essential to build a human friendly product/service. In this paper we proposed a Mobile voting application in Kuwait which allows voters to cast their vote and engage in different surveys without fear of complicated verification procedures. The surveys are the cornerstone of marketing, and for reliable and trustworthy outcomes, the project incorporates experience of voting system into a smart and simple survey mobile application, and will be compatible with all mobile platforms. It will be designed to be ease of use and customizable, so any one can use it for their own survey, also it will eliminate the duplication of votes to get more credibility. The key features of the application include the use of ID and GPS as main aspects to authenticate the user account, and provide the necessary security measures by broadening the device's scope. GPS will be the main contribute as it will be used in a new technique. The voting application is based on the three-step process used by each survey application, i.e. user authentication, voting, and results assessment.
Keywords:	Survey Mobile App, GPS, Mobile Voting System, Smart Voting, Social Influence
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	INVESTIGATION OF FAULT TOLERANT SCHEDULING ALGORITHM FOR TASK IN CLOUD SYSTEMS USING ANT COLONY OPTIMIZATION
Author:	SRIDEEPA.T, DR.BABYDEEPA.V
Abstract:	Cloud computing is defined as the sharing of computing and communication resources across a network of distributed data centers and is used by a wide range of systems. Fault tolerance is defined as a device's ability to gracefully respond to a sudden software or hardware failure. The addition of fault tolerance in cloud computing has several advantages, including improved failure recovery, lower infrastructure costs, and increased normal overall performance measurements. The ability to continue operating in the event of a power outage is the lowest level of fault tolerance. Many fault tolerant computer systems replicate all operations as a way to increase reliability by performing the same work on or more duplicate systems, so that if one fails, the other can take over. So, theoretically, fault tolerance techniques are used to anticipate disasters and take appropriate action before they occur. So, in a way, fault tolerance ideas are used to anticipate these disasters and take appropriate action before they occur. In this paper, we planned a set of rules as well as an aco to dynamically schedule appear real-time tasks with useful resource and fault-tolerant requirements directly to multiprocessor systems. They quantify the effectiveness of each of those strategies in improving the assure ratio, which would be defined as the percentage of general tasks completed within the system within the time limits. Our proposed Fault tolerance scheduling primarily based totally on heuristic algorithms has ambitions at reaching every fault tolerance and immoderate beneficial aid usage in the cloud. The experimental results show off that in comparison with Existing Dynamic Fault Tolerance Scheduling Techniques (DFTST) and Proposed Fault Tolerance Scheduling with Ant Colony Optimization (FTSACO) Algorithms extensively improves the common scheduling overall performance, achieves a more diploma of fault tolerance with immoderate beneficial aid usage, minimizes the advise reaction time and task rejection ratio, and reduces energy consumption.
Keywords:	Cloud Computing, Fault Tolerance Scheduling, Host Active Time, Genetic Algorithm, Ant Colony Optimization, Energy Consumption
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	EXPERIMENTAL INVESTIGATIONS TO FACE RECOGNITION USING OPENCV FOR MONITORING TURNOUT TIME
Author:	Dr. J. Ravindra Babu, Mr. Tata Balaji, Dr. Surya Prasada Rao Borra, Dr.S. Hrushikesava Raju, M. Venkata Subba Rao, Dr. A.Geetha Devi, Dr K Koteswara Rao
Abstract:	Face Recognition is becoming an increasingly important technology in the present world. There are various types of biometric identifications including DNA, fingerprints, signature recognition, hand geometry, palm print, etc. For this type of recognition, some action has to be done by the user like placing a finger on the machine to detect. While, Face Recognition does not require any user actions. So, we can say that face recognition is one of the most successful biometric identification methods. We have been already engaging with facial recognition in our daily life without even realizing it. Face recognition plays a vital role in fields such as identifying the retail crimes, finding missing person, to help the blind, protects law reinforcement, aids forensic investigations, to diagnose diseases, unlocks smart phones, facilitates secure transactions, validates identity at ATMs, control access to sensitive areas and a lot more. The challenge that can be encountered in face recognition is to detect the face from single image that is stored in the database. Face detection is a challenging task as the faces are not rigid and they will be changing in size, shape, colour, etc. Face detection become more challenging task when given image is not clear and not containing a proper lightning, not facing camera etc. In this work we are building a face recognition system for monitoring turn out time using OpenCV. A facial recognition system is a computer application. This technology is used for identifying or verifying a person from a digital image. This digital image can be obtained from a video frame or from live images detected by the camera/webcam. Facial recognition involves two stages. First we have to detect the face. For this process, a photo is searched to find a face, after finding the face in the image, the image is processed to crop and extract the person’s face in a square box. And the second phase is Face Recognition, where the face detected by the above process is compared with the images in the dataset, to decide who that person is. In this work, we are using the HOG (Histogram of Gradients) algorithm. A Face can be detected by cropping the face and removing the noise using the HAAR cascade classifier. Later the features of the face can be extracted by using the HOG extractor. Then we will run an SVM model for face recognition. The results show that the system can recognize the faces captured automatically by the camera are accurate and efficient
Keywords:	Open CV, Face Recognition. Monitor
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	PROPER NAMES KNOWLEDGE BASE FOR INTELLIGENT MOBILE APPLICATION
Author:	GAZIZA YELIBAYEVA, LAURA ORYNBAY, GULMIRA BEKMANOVA, AYAULYM SAIRANBEKOVA
Abstract:	This article proposes a semantic knowledge base of proper names of the Kazakh language with about 100 semantic features. A semantic knowledge base of proper names of the Kazakh language allows obtaining new scientific results in expanding the functions of the Kazakh language, developing information resources in digital format and scientific-linguistic foundations of the Kazakh language for intelligent information systems with different capabilities. The analysis showed that Kazakh names have a large number of semantic features from the origin of names to their emotional coloring, borrowing from other languages, combining different roots, and so on, which are included as properties of the semantic knowledge base. The results of the work will be used to develop software application “Fascinating Onomastics” that will allow anyone who wants to do the analysis of names by origins, distributions, and variations as well as choose a name according to the given parameters. Moreover, this research contributes to the extraction and classification of proper names in many applications in the field of natural language processing and text analysis. By extracting proper names from the text, we provide a POS tagging information with semantic features. The study is carried out within the framework of the project BR11765535 “Development of Scientific and Linguistic Foundations and IT Resources to Expand the Functions and Improve the culture of the Kazakh Language” and upon completion of the project, the mobile application will be published on the Google Market and the Apple Store.
Keywords:	Fascinating Onomastic, Semantic Features, Knowledge Database, Logical Rules, Artificial Intelligence.
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	AN EFFICIENT ATTACK DETECTION FOR INTRUSION DETECTION SYSTEM (IDS) IN INTERNET OF MEDICAL THINGS SMART ENVIRONMENT WITH DEEP LEARNING ALGORITHM
Author:	FATIMAH SALEEM ABDULKAREEM, NOR FAZLIDA MOHD SANI
Abstract:	Recently, the Internet of Things (IoT) has been an invention for the creation of intelligent worlds. IoT is considered a widely recognized implementation that includes intelligent health care, intelligent transport, and intelligent grids. In any technology depending on the IoT model, in which the Internet of Medical Things (IoMT) is an important technique, privacy and secrecy are considered the major problems driven by numerous attacks triggered by intruders. The detection of unknown attacks is one of the main challenges in intrusion detection system (IDS). Researchers have performed multiple typing and detected anomaly traffic methods in the past decades without earlier understanding the attack signature specifically to the IoT environment. Therefore, an intrusion detection method for attacking and detecting anomalies in an IoT system must be enhanced. To achieve this, we measured the performance of three deep learning algorithms for normal and abnormal detection of IDS, and a comparison was made to select the best performance of the deep learning algorithm for detection in IDS, such as RNN, DBN and CNN. The CICIDS2017 dataset was used to analyze the performance of the existing intrusion detection system model. Additionally, the results of the deep learning algorithms will be evaluated using five confusion matrices, namely, accuracy, precision, recall, F1Score, and false-positive rate). It should be noted that the results showed a good average because most of them exceeded 90% of the total confusion matrix for all three deep learning algorithms that have been evaluated.
Keywords:	IoMT, Intrusion Detection, Anomaly Detection, Deep Learning.
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	EMPIRICAL INVESTIGATIONS TO SENTIMENT ANALYSIS OF MOVIE REVIEWS USING LSTM
Author:	DR. HABIBULLA MOHAMMAD, DR. J. RAVINDRA BABU, DR. SURYA PRASADA RAO BORRA. DR.SRINU PYLA 5MR.TATA BALAJI, MRS. B. MOUNIKA, KOTESWARA RAO KODEPOGU
Abstract:	Sentiment analysis or opinion mining is the computational study of people’s opinions, sentiments, attitudes and emotions expressed in written language. It is one of the most active research areas in Natural Language Processing in the recent years. Sentiment analysis aids corporations in making decisions and changes in their business or service models based on the feedback of the customers regarding the current models. Most sentiment analysis problems are classification problems (positive/neutral/negative) and not regression problems. It comes under Sequential problems which are a class of problem in machine learning where the order of the features presented to the model is important for making predictions. In this project, we study the existing classification model based on Recurrent Neural Network (RRN), build a machine learning (Classification) model using long short-term memory (LSTM) network to overcome the Vanishing Gradient problem faced in RRN. The model takes an IMDB movie review dataset with 50,000 reviews as input; trains on 25,000 and uses the experienced acquired so far to classify another 25,000 reviews into positive and negative categories. We animate the results of the model using graphs.
Keywords:	LSTM, RNN, Sequential Problems, Sentiment, Movie Reviews
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	BUY NOW PAY LATER SERVICES ON GENERATION Z: EXPLORATORY DATA ANALYSIS USING MACHINE LEARNING
Author:	YOSY ARISANDY, YOSZA BIN DASRIL, SHAHRUL NIZAM BIN SALAHUDIN, MUCH AZIZ MUSLIM, ARISMAN ADNAN2, GOH KHANG WEN
Abstract:	The buy now, pay later (BNPL) business model is an innovative approach to installment loans. It allows customers to take immediate possession of their purchase, with or without a down payment. Furthermore, the majority of BNPL loans are set up to require four payments. However, this type of loan comes with its own set of risks and challenges. This article examines the risk of BNPL as a product for consumers known as Generation Z. The data used is secondary data provided by Kaggle in csv format (loan data.csv) contains 159,584 postpaid customer records and 28 features analyzed through descriptive and Exploratory Data Analysis (EDA). The results show that the majority of pay later clients are married and known as millennials are the ones who used pay later services the most (52.10%). Generation Z has the greatest rate of loan defaults which is about 34.16% with the time employee is about 0-8 months (35.8%). Furthermore, the results indicated that the unemployed generation Z has the highest default percentage of 32.16%. This Exploration data analytic is viewed as a step towards gaining a better understanding of consumers so that predictions, suggestions, and recommendations can be made for potential customers and market paylater segmentation to find the right target market, thereby positively impacting company profits.
Keywords:	Buy Now Pay Later, Risky, Generation Z, Exploratory Data Analysis, Machine Learning
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	BUILDING A NEURAL NETWORK TO ASSESS THE LEVEL OF OPERATIONAL RISKS OF A CREDIT INSTITUTION
Author:	EKATERINA VITALEVNA CHUMAKOVA, DMITRY GENNADIEVICH KORNEEV, MIKHAIL SAMUILOVICH GASPARIAN, ANDREY ALEKSANDROVICH PONOMAREV, ILIA SERGEEVICH MAKHOV
Abstract:	This article considers the issues of managing the operational risks of a credit institution arising in the process of using information technologies. To manage operational risks associated with the use of IT in banks, the authors propose methods based on the use of artificial neural networks. The decisive stage in operational risk management is the collection and intellectual analysis of data. At this stage, hazards become risks that can be implemented into the logic of managerial decision-making. Optimal IT risk management involves online monitoring of numerous parameters that affect the possibility of risks and determine their consequences. The risks that banks face using IT solutions extend to third-party IT providers that many banks rely on for cloud storage and other services. These systems can slow down or fail, preventing customers from accessing ATMs or mobile apps. Even the speed of a technological change creates operational risk. In connection with the above, the automation of operational risk management based on the use of intelligent technologies is one of the most urgent tasks for credit institutions. The authors suggest that the networks can be trained on statistical data specific to each institution to enable accurate forecasting and event analysis using complex neural network technologies. The study identified models that demonstrated accurate results (it is necessary to take into account some limitations stated by the authors) on the training set formed by experts, leading to an expected increase in forecast accuracy by at least 40% upon implementation. The practical recommendations offered in the study have the potential to improve risk management practices and enhance the efficiency of credit organizations.
Keywords:	IT Risks, Artificial Neural Network, Machine Learning, Feedforward Neural Network, Keras High-Level Library (Framework).
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	A REAL-TIME HYBRID-YOLOV4 APPROACH FOR MULTI-CLASSIFICATION AND DETECTION OF OBJECTS
Author:	SMITA RATH, SUSHREE BIBHUPRADA B. PRIYADARSHINI, DEEPAK KUMAR PATEL, NARAYAN PATRA, PRABHAT SAHU
Abstract:	This paper improves the object detection accuracy for detecting objects in complex scenes and ensures real-time classification operations by planning a novel detection method called lightweight and efficient hybrid YOLOv4 model. In this context, Computational vision is one of the most useful and entertaining forms of artificial intelligence (AI) used in everyday life. Computer vision study focused on replacing intricate aspects of the human world with sophisticated AI and computers. Deep neural networks have recently become an essential part of several industries due to their renowned ability to handle visual input. One of the main directions that computer vision has taken is the domain of classification & tracking of objects employing neural networks, which are presently being employed by relevant trendsetting enterprises specializing in solving several arrays of predicaments such as security, health care, and agriculture. The main factors affecting the development of computer vision are the volume of data it generates, as well as the amount it utilizes to train and enhance it. In this paper, a method for categorizing and detecting objects utilizing an object detection algorithm namely hybrid-YOLOv4 is proposed. Convolutional neural networks provide extremely accurate object tracking and feature extraction out of the images. Strategies such as Bag-of-Specials and Bag-of-Freebies are used in item identification and DarkNet is used in the backbone that increases the feature exchange and reutilization. Thus, the improved network design maximizes both identification accuracy and speed. Additionally, two new extra blocks in the neck and backbone enhance feature extraction and reduce processing expenses. The model was compared with other object detections methods. According to the experimental findings, mean average precision (MPS) of YOLOv4-hybrid model was found to be 0.986 better than that of YOLOv4 and other object detection models.
Keywords:	Artificial Intelligence, Computer Vision, Faster RCNN, YOLOv4, RCNN
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	CROPS FINE CLASSIFICATION IN HYPERSPECTRAL IMAGERY BASED ON PRINCIPAL COMPONENT ANALYSIS (PCA) AND DEEP LEARNING
Author:	S. JAMALAIAH, Prof. K. MANJULA VANI
Abstract:	Hyperspectral image classification has been deployed in a number of real-world scenarios, such as agricultural, quality control of agro-food products, and medical fields. Hyperspectral classification is difficult due to inter-class similarities, variability, nested regions, and overlaps. For the classification of hyperspectral data, traditional convolutional neural network (CNN)-based algorithms are applied. CNN primarily employs two-dimensional (2D)-CNN for hyperspectral extracting features, leaving intra band correlation of hyperspectral images ignored. The three-dimensional (3D)-CNN extracts the features by combining spatial-spectral bands of hyperspectral images, which is a complicated model, in order to accomplish this performance. This study proposes a customized 3D-CNN that uses hyperspectral images to integrate both spatial and spectral information. The proposed method first reduces the data by using dimensionality reduction, generates the image cube by using image patching, applies the customized 3D-CNN and, using a softmax classifier, classifies different crops in suburban areas using high-resolution features acquired by the customized 3D-CNN. The customized 3D-CNN model produces maximum overall accuracy of 99.88% and 99.83% for Indian pines and paviaU datasets. The model produces accurate results that are verified against existing solutions. The performance of the proposed method is validated using benchmark hyperspectral datasets from the study area.
Keywords:	Classification, CNN-Convolution Neural Network, Convolution; Deep Learning (DL), Deep Neural Network (DNN), FCL-Fully Connected Layer, HSI-Hyperspectral Image, NN-Neural Network, PCA-Principal Component Analysis, 3D-CNN-3-Dimensional Convolution Neural Network.
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	FACTORS INFLUENCING EMPLOYEES' INTENTION TO PARTICIPATE IN A BRING YOUR OWN DEVICE IN THE PORT SUPPLY CHAIN NETWORK: A CORRELATIONAL STUDY USING UTAUT2 THEORETICAL
Author:	BOISON, DAVID KING , ANTWI-BOAMPONG, AHMED2AUGUSTINE, BLAY, , DOUMBIA MUSAH OSUMANU4 ASIEDU ESTHER, SARBENG, KWAME OWIREDU
Abstract:	The study aimed to assess the factors that influenced port users' willingness to participate in BYOD programs in Ghana's Maritime and Port sector. The extended Unified Theory of Acceptance and Use of Technology (UTAUT2) was used as the theoretical framework for the quasi-quantitative study. The study examined whether eight factors were predictors of the intention of Ghanaian employees to participate in a BYOD program, moderated by social influence. The study used principal component analysis (PCA) in SPSS and structural equation modeling in Stata to analyze and report the data. The results showed that only three factors, namely Performance Expectancy (PE), Effort Expectancy (EE), Facilitating Conditions (FC), and HT, significantly influenced employees' behavioral intention (BI) to participate in a BYOD program, while Social Influence (SI), Hedonic Motivation (HM), and Price Value (PV) had no effect on Behavioral Intention (BI). Age did not moderate the influence of any factor on BI. The study provides insights into the port supply chain network's usage of BYOD and will aid academics in explaining the discrepancies between the UTAUT2 theoretical framework's predictions for different industries and specialties. The study's findings will also be useful for researchers who aim to implement the UTAUT2 theoretical framework to understand employees' BI to join the BYOD program in any industry. From a practical perspective, the study will assist managers in the port business in Ghana and the sub-region in focusing on the important structures that constitute the initial steps to introducing BYOD in the port supply chain industry.
Keywords:	BOYD, Maritime Ports, Ghana, Supply Chain Network, UTAUT2
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	A SECURITY FRAMEWORK PROTECTING VIRTUAL MACHINES AGAINST ATTACKS ON MIGRATION AND PERSISTENCE IN CLOUD COMPUTING ENVIRONMENT
Author:	S. MAHIPAL, V. CERONMANI SHARMILA
Abstract:	In a cloud computing environment, Virtual Machine (VM) migration achieves energy efficiency, efficient resource management, and load balancing. VM persistence is another area that leads to increased performance. However, both of them do have security vulnerabilities. There are existing approaches followed by Virtualization technology vendors. However, there is a need for further research to leverage security in the aforementioned areas. Towards this end, in this paper, we proposed a security framework that ensures that VM migration and VM persistence occur without causing cyber-attacks. The framework has two algorithms proposed to realize this objective. Safe Virtual Machine Migration (SVMM) is meant for protecting VM from VM hopping attacks on the VM migration process while another algorithm known as Safe Virtual Machine Persistence (SVMP) focuses on preventing attacks on VM persistence. Both mechanisms are crucial for leveraging cloud performance and Quality of Service (QoS). The proposed framework is realized with a simulation study using CloudSim. The experimental results showed that the proposed approach is capable of handling attacks while VM is being migrated and when VM is being persisted.
Keywords:	Cloud Computing, VM Migration, VM Persistence, Secure VM Migration, Secure VM Persistence
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	ANALYSIS: INHIBITING FACTORS OF TACIT KNOWLEDGE SHARING FROM LEADERS TO NEW EMPLOYEES ON INTERNET COMPANY
Author:	MUHAMMAD RAJENDRAYAN DANESWARA, SFRENRIANTO
Abstract:	There are some common knowledge management problems in a company specifically on tacit knowledge sharing where tacit knowledge is highly personal and difficult to formalize, which makes it difficult to communicate or share it between leaders and new employees. In this case, this study will go directly to one of the companies in the field of internet service providers to analyze the real problem and find what are the inhibiting factors that hinder the knowledge sharing from leaders to new employees. This research uses a survey distribution method with non-interactive questionnaire techniques out of 104 respondents and uses a literature review method to obtain theories. This study also includes a literature review by citing various theories and statements from several previously conducted studies to support and help find factors that become obstacles to knowledge sharing. The results of this study indicate that are several factors but lack of time for knowledge sharing is the main factor that inhibits the knowledge sharing between leaders dan new employees.
Keywords:	Knowledge Management, Tacit Knowledge, Knowledge Sharing, Leaders, New Employees
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	MEASURING THE ACCURACY OF SEARCH INTERVAL PARAMETERS ON RIDGE POLYNOMIAL NEURAL NETWORK IN EARLY DETECTION OF BRAIN CANCER
Author:	RIAH UKUR GINTING, POLTAK SIHOMBING, SYAHRIL EFENDI2, AMILA, BURHANUDDIN DAMANIK
Abstract:	Brain cancer is a disease that causes the highest death for men and women at the age of 20-30 years. Epidemiology of brain cancer data in Indonesia to date is inadequate, this is due to suboptimal diagnostic techniques and incomplete case registration problems. One of the factors causing delays in early detection of brain cancer is the high cost and lack of public knowledge about the risk of brain cancer. The purpose of this paper is to develop a method and system that is able to detect brain cancer early using a polynomial neural network ridge algorithm based on Accuracy of Search Interval Parameters. Ridge polynomial neural network is one algorithm with good accuracy results for early detection of brain cancer from artificial neural network methods. This research will use eight input variables including: headaches gradually becoming more frequent and more severe, nausea and vomiting without cause, impaired memory, seizures, tingling and numbness in the arms and legs, visual disturbances such as blurred vision, related problems with the sense of hearing and impaired balance (difficulty in moving). The data will use weights from genetic algorithms arranged in time-series and then trained using artificial neural networks with the polynomial neural network ridge algorithm. The results of early detection of brain cancer will be visualized in patients at Haji Adam Malik General Hospital Medan which was shown in the 427th iteration by achieving an MSE of 0.021844.
Keywords:	RPNN, Search Interval Parameters, MSE, Brain Cancer.
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	AN ENHANCED APPROACH FOR TEST SUITE REDUCTION USING CLUSTERING AND GENETIC ALGORITHMS
Author:	SARAH M. NAGY, HUDA A. MAGHAWRY, NAGWA L. BADR
Abstract:	Software testing is a procedure used to evaluate the quality, accuracy, and completeness of a generated computer software. It entails a series of actions taken with the goal of identifying software faults so they can be fixed before the product is made available to end users. Testing a program against a collection of inputs known as test cases is one of the most practical ways to find faults in it. Redundant test cases are useless. Besides, they increase the testing effort, testing costs, and testing time. Testing involves spending a lot of time on a lot of unreliable test cases. An excessive cost is wasted when redundant or outdated tests are run that do not increase fault detection capabilities. In this study, the objective is to propose an enhanced approach for test suite reduction to enhance the regression testing process. This is achieved by reducing the time spent in testing by finding a subset of test cases that fulfill the requirements and discovering most of the faults already present. This subset is known as a reduced test suite. A test suite is a set of tests that enables testers to run and report the status of the test execution. Therefore, a clustering-based approach is proposed to considerably minimize the test suite. The proposed approach applies the K-means++ clustering algorithm. Utilizing K-mean++, test cases are grouped into groups depending on their degree of similarity. Then, a multi-objective genetic algorithm is applied to reduce the test suite in each cluster based on code coverage. For any unsupervised clustering algorithm, determining the optimal number of clusters into which the data can be divided is a crucial step. Therefore, two methods were experimented to determine the optimal k: elbow method and silhouette analysis method. The proposed enhanced approach outperformed previously published approaches in terms of test suite reduction and code coverage rate.
Keywords:	Regression Testing, Clustering, Genetic Algorithm, Test Suite Reduction
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	PROPOSE A MODEL OF CONTINUANCE INTENTION TO USE IOT SMART HOME IN MALAYSIA
Author:	SITI FARAH HUSSIN1, MOHD FAIZAL ABDOLLAH, IBRAHIM AHMAD
Abstract:	The Internet of Things technology for smart home (IoTsh) are homes equipped with communication networks, systems, and sensors that can be used for controlling, monitoring, and scheduling enabling hardware IP devices according to user needs. The use of IoTsh can improve self-management, improve social care systems, and provide a better lifestyle. Although there are many IoTsh advantages, existing studies only focus on IoTsh acceptance and adoption. Nevertheless, studies regarding the intention to continually use IoTsh remain scarce, which calls for further investigation. To address this gap, this study proposes an integrated model for the purpose of understanding the intention of continuous usage of IoTsh among Malaysian users, as well as for investigating factors that impact or prevent the continuance intention of using IoTsh. The model combines three information system theories as the underlying theories, namely the Unified Theory of Acceptance and Use of Technology 2 (UTAUT2), Expectation-Confirmation Model (ECM), and Hofstede’s theory. The proposed model provides important insights for IoTsh providers, manufacturers and governments on continuance intention to use IoTsh, understanding the users' desires and subsequently designing their services.
Keywords:	Continuance Intention, ECM, Hofstede, Iot Smart Home, UTAUT2
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	OPTIMIZING SENTIMENT ANALYSIS OF AMAZON PRODUCT REVIEWS USING A SOPHISTICATED FISH SWARM OPTIMIZATION-GUIDED RADIAL BASIS FUNCTION NEURAL NETWORK (SFSO-RBFNN)
Author:	P. RADHA, Dr. N. SUDHA BHUVANESWARI
Abstract:	Sentiment analysis automatically identifies and extracts subjective information from text, which can help understand people’s opinions, emotions, and attitudes towards a particular topic. Sentiment analysis has become increasingly important in recent years, as online reviews and social media have become popular platforms for people to share their opinions and experiences. However, there are several challenges in sentiment analysis, including the complexity and ambiguity of language, the lack of context, and the cultural and linguistic differences that can affect the interpretation of sentiment. These challenges can result in inaccurate sentiment analysis, which can have negative consequences, such as misleading product reviews or biased customer feedback. In this paper, “Sophisticated Fish Swarm Optimization (SFSO)-guided Radial Basis Function Neural Network (RBFNN)” is proposed to perform sentiment analysis with enhanced classification accuracy. The SFSO algorithm optimizes the parameters of the RBFNN, which enables the model to adapt to varying review topics and sentiments. The SFSO algorithm’s ability to explore the search space of the RBFNN parameters results in improved accuracy and performance in sentiment analysis. The proposed approach was evaluated on a dataset of Amazon product reviews and compared to other state-of-the-art sentiment analysis techniques. The proposed approach has potential applications in Books, Kindle stores, Tools and Home Improvement, and Industrial and Scientific domains where sentiment analysis is critical for understanding customer opinions and feedback. The results demonstrate that the proposed approach outperforms state-of-the-art classifiers in terms of Classification accuracy, F-measure, Fowlkes-Mallows Index, and Matthews Correlation Coefficient.
Keywords:	Sentiment Analysis, Amazon, Optimization, Classification, Radial Function, Fish Swarm.
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	METAVERSE: NEW WAYS STUDENTS WILL INTERACT IN FUTURE LEARNING
Author:	HANSEN TANJAYA, VIANY UTAMI TJHIN
Abstract:	This research aims to evaluate the use of metaverse in learning systems at universities, with a focus on the Gather Town platform. This research was conducted using interview techniques with 10 Bina Nusantara University students and analyzing data using the NVivo software. The results showed that students responded positively to the use of Gather Town in learning, considering that this platform provided a more interactive and enjoyable environment compared to traditional learning methods. However, students also stated that technical factors such as internet connection and device compatibility must be considered. In addition, several students stated that Gather Town should be used as an additional means of learning, not as the only means. From the results of this study, the use of metaverse can be an effective alternative in the learning system at universities. However, technical factors must be considered and used as an additional means. In addition, the results of this study also show that Gather Town has several features that are useful in learning, such as the breakout room feature, which allows students to interact with other students in a more focused manner, and the polling feature, which allows lecturers to evaluate students' understanding of the material being taught. Lecturers can also use the presentation feature to present material more interactively. This research suggests that universities can evaluate the potential of using Gather Town and other metaverses in their learning systems. Overall, the results of this study indicate that the use of Gather Town in the learning system at universities can increase the interactivity and quality of learning and be a good alternative to conventional learning.
Keywords:	Gather Town, Metaverse, Education, University, Perception
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	A SURVEY ON DETECTION OF SEVERITY IN DIABETIC RETINOPATHY USING MACHINE LEARNING MODELS
Author:	N V NAIK , HYMA J , PVGD PRASAD REDDY
Abstract:	Diabetes has rapidly become a leading cause of death globally. The metabolic irregularities and complications of diabetes, such as high blood sugar and insulin production, cardiovascular diseases, nephrological problems, neurological disorders, and diabetic retinopathy (vision loss), all have their origins in the disease itself. Diabetic retinopathy (DR) refers to a serious problem faced by diabetic affecting retina. It causes leakages in blood vessel walls within retina, thereby damaging it. It is considered the major cause of blindness because of its rapid onset and absence of symptoms. In order to properly intervene and treat DR, it is crucial to be aware of the early clinical indications of the disorder. Therefore, it's important to get your eyes checked often so you know whether you need to see an eye doctor right away so you can prevent any irreversible damage to your eyesight. The goals of this study are to provide a critical evaluation of many machine learning and deep learning methodologies and observe patterns to enhance strategies of current work, to highlight obstacles, and to recommend prospective future research directions. The results can help shape future research agendas, and the recommendations can help shape future models for diabetic retinopathy algorithms that have both high generalizability and high performance.
Keywords:	Diabetic Retinopathy; Machine Learning; Deep Learning; Blood Vessels; Taxonomy
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	AN IMPROVED EM-CNN DEEP NEURAL NETWORK FOR CLASSIFYING MILD-DIABETIC RETINOPATHY FROM NORMAL IMAGES
Author:	EBIN PM, D. P RANJANA
Abstract:	One of the main issues middle-aged people have because of uncontrolled blood sugar is vision impairment, which is also known as diabetic retinopathy (DR). The early signs of diabetic retinopathy are minor abnormalities in the retinal capillaries known as microaneurysms and intraregional hemorrhage. Due to the lack of resources and skilled medical professionals, clinical diagnosis of diabetic retinopathy is delayed and challenging process, making early detection even more crucial to prevent the spread of the disease. Herein lies the value of an automated DR detection system to spot the early signs of DR. The research work uses a combination of two types of fundus images contrast-limited adaptive histogram equalization (CLAHE) and non-CLAHE. CLAHE is a technique used to enhance the contrast of an image by adjusting the intensity distribution, while non-CLAHE images are unaltered. Using a binary classification approach, the researchers in this publication constructed a new model called Experimental Minimal Convolutional Neural Network (EMCNN) model to categorize Mild-DR and No-DR fundus pictures. By training the EM-CNN model on both types of images, the researchers aim to improve the accuracy of DR classification. The result is compared with the model, which already exist. The model achieved 98%accuracy, which is better than existing models.
Keywords:	CLAHE, Convolutional Neural Networks, Deep Learning, Diabetic Retinopathy, EMCNN.
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	COLLABORATIVE VISUALIZATION FRAMEWORK FOR CROSS-FIELD WORKING GROUP: ANALYSIS THROUGH SMART PLS
Author:	DANIAL ILMAN MUHAMMAD HASNI, ALIZA SARLAN
Abstract:	This study is conducted to develop a collaborative visualization framework in the cross-field working group. The goal of this project is to provide a proper framework that can be used to develop a platform to allow collaborative visualization to be implemented inter-disciplinary groups, in two different settings: university students and research groups in research and development companies and institutions. The study begins with preliminary works to define the collaborative visualization published in the previous research. It focuses on the factors to develop an effective collaborative working environment through visualization and shared understanding among the staff/users from inter-disciplinary backgrounds. In addition, this study also investigates the interaction between human cognition, collaborative factors, and ICT attributes of visualization in developing an efficient working group to achieve common goals and objectives. To conclude, the framework will be tested to validate its possible contributions to the targeted collaborative working groups. The study is hoped to contribute to the identification of factors that connect the application of collaborative assisted tools in visualization with the development of human cognition and shared understanding towards achieving efficiency especially in a multi-disciplinary working environment.
Keywords:	Cross-Field Working Group, Collaboration Visualization, Human Cognition, Structural Equation Modelling, Smart-Pls
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	ANALYSIS ROLE OF ROBOTIC PROCESS AUTOMATION IN ACCOUNTING AND BUSINESS
Author:	MEIRYANI, ALFIA DAVA ZAHRA, FELICIA CHRISTINE CHANDRA, DEZIE LEONARDA WARGANEGARA
Abstract:	Robotic Process Automation (RPA) is a revolution in automation technology that can improve company competitiveness. The purpose of this study is to provide theoretical and empirical evidence of the benefits and role of robotic process automation in accounting and business. The method used in this research is descriptive research method, where this research collects detailed data from various literatures. Robotic Process Automation (RPA) is software or technology that enables software to perform business processes efficiently and quickly. So as to reduce worker errors, research results show that robotic process automation (RPA) is a software or system that can help companies in accounting and business, such as sending messages to many customers using technology/systems. Decision Minister of Industry and Trade Number 121/MPP/Kep/2/2002 which explained that the financial statements. The company must be audited. Based on the regulation, the auditor will remain survive the emergence of RPA.
Keywords:	Robotic Process Automation, Accounting, Business, Decision Making
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	HYBRID SOLAR POWER GENERATION PREDICTION USING SUPPORT VECTOR MACHINES AND K-NEAREST NEIGHBORS OPTIMIZED BY DEEP LEARNING TECHNIQUES
Author:	Mrs. K.V.B. SARASWATHI DEVI,Dr. MUKTEVI SRIVENKATESH
Abstract:	Solar power is one of the world's most popular and fastest-growing sources of clean energy. Nevertheless, it relies on sunshine, which is a finite natural resource. Power production predictability is essential for integrating photovoltaic (PV) systems into the grid, and this is especially true for solar photovoltaics. All across the world, PV systems are used to generate solar power. Solar power sources are unpredictable and uncontrollable since the output of the PV systems is intermittent and highly reliant on environmental conditions. These include irradiance, humidity, PV surface temperature, and wind speed. Photovoltaic power generation is highly unpredictable, so it is essential to prepare ahead for solar power generation as in solar power prediction is required for the electric grid. As renewable power is weather-dependent and prone to uncertainty, this forecast is difficult to anticipate accurately. Some of the environmental factors that affect a PV system's power output are explored. With the use of Machine Learning (ML) algorithms, it is possible to predict the amount of power that will be generated based on the meteorological conditions. An ensemble of machine learning models was utilized in this work to improve the model's accuracy. In this research, an Integrated Support Vector Machine with K-Nearest Neighbor (ISVM-KNN) model is proposed for prediction of solar power generation. Simulated findings reveal that compared to current approaches, the suggested method has a lower placement cost. It was found that the proposed ensemble model outperformed the traditional individual models when compared to a standard model that included all of the combination procedures.
Keywords:	Photovoltaic, Solar Power, Machine Learning, Weather Conditions, Support Vector Machine, K-Nearest Neighbor, Ensemble Approach.
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	MATURITY CLASSIFICATION OF TOMATOES USING CONVOLUTIONAL NEURAL NETWORK
Author:	OCTAVIANUS JAMES MARSHAL, RIYANTO JAYADI
Abstract:	Tomato is a fruit used daily that requires good quality, and quality is also an essential factor in sales percentage. In the market. Currently, the selection of quality tomatoes is still made mainly by humans, with several areas for improvement, such as the accuracy and consistency of the results obtained due to limited human perception. With ever-improving technology, it is now possible to train computers to classify images based on specific characteristics. This study proposes a classification model to classify tomato images using Convolutional Neural Networks (CNN). A total of 300 pictures of tomatoes have been selected from 480 pictures taken using a smartphone camera, and these images will be divided into three classes, unripe, ripe and rotten. Each class consists of 100 images and will be divided into 70% as training data, 15% as validation data and 15% as test data. In this study, we compared the accuracy of the VGG19 fine-tune and unblock layer models. And will compare the kernel used in the VGG19 model to determine the impact of the kernel on the classification accuracy that increases the model kernel parameters. From this study, CNN can see that the parameters used can affect accuracy.
Keywords:	Classification, Neural Network, Tomato, Maturity
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	A SURVEY ON USED VEHICLE PRICE ESTIMATION SYSTEMS USING ARTIFICIAL INTELLIGENCE METHODS
Author:	THURAYA ALNAJIM, NOUF ALSHAHRANI, OMAR ASIRI
Abstract:	Recently, due to the high cost of new cars, which most buyers cannot afford, the market for used car transactions in the Kingdom of Saudi Arabia has been rapidly expanding. Several e-commerce websites offer intermediary services between buyers and dealers of used automobiles. However, it is very helpful to have information about the correct vehicle price for many buyers and sellers before making any decision about selling or buying a used car. Therefore, there is a great demand to develop an accurate vehicle price estimation system through the employment of machine learning (ML) and deep learning (DL) approaches. However, a large number of significant factors influence the price of a vehicle, making vehicle price estimation a challenging task. In general, the standard regression methods might not be efficient for high-dimensional data. This paper aims to investigate the recently developed vehicle price estimation systems that are based on the employment of ML and DL approaches. The recently developed systems are discussed and criticized in detail. In addition, we present a set of evaluation metrics to assess the efficiency of any vehicle price prediction system.
Keywords:	Vehicle Price Estimation, Deep Learning, Feature Extraction, Machine Learning, Used Cars.
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	SPECTROGRAM FLIPPING: A NEW TECHNIQUE FOR AUDIO AUGMENTATION
Author:	QASEM M. M. ZARANDAH, SALWANI MOHD DAUD, SAMY S. ABU-NASER
Abstract:	Data augmentation is a technique used to increase the amount and diversity of training data in deep learning models. In this paper, we propose a new audio data augmentation technique that combines traditional audio augmentation methods such as time-stretching, pitch-shifting, and noise injection with a novel technique called "spectrogram flipping." Spectrogram flipping involves taking the spectrogram of an audio signal, flipping it horizontally, and then converting it back to a time-domain audio signal. This technique results in audio data that is both diverse and realistic. We evaluate our proposed technique on a repository diseases classification task using a deep neural network. Our experiments show that our technique improves the accuracy of the classification task model compared to traditional audio augmentation methods. We also show that our technique is computationally efficient and easy to implement. Overall, our proposed audio data augmentation technique is a valuable addition to the toolbox of deep learning researchers working with audio data. It has the potential to improve the performance of a wide range of audio-based deep learning models.
Keywords:	Data Augmentation, Audio, Spectrogram Flipping, Deep Learning
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	A FULL-SCALE ANALYSIS ON CHALLENGES AND ISSUES OF NEXT GENERATION (5G) COMMUNICATION IN HETEROGENEOUS WIRELESS NETWORK BASED ENTERPRISE APPLICATIONS
Author:	SUBHRA PROSUN PAUL , D. VETRITHANGAM
Abstract:	In todays globalized world, there is a continuously growing interest in next-generation (5G) communication of versatile wireless network services for its popularity. A great technological transformation has been made in next-generation (5G) communication in cellular phone services in terms of screen size, data processing capacity, resolution density, and cost. To develop network coverage area, proper energy and bandwidth utilization, and faster communication with a cheap rate, next- generation (5G) communication has been introduced at several interconnected communication levels. No doubt, we still have to handle some remarkable challenges like the cell internal interference problem of a heterogeneous network, appropriate implementation of software-defined network concept at network architecture stage, network storage, resource management, and security in this 5G communication area. Additionally, upgraded signal processing, perfect channel estimation, network optimization, and successful mobility management are the important challenges to be faced in this field. Our main aim is to identify the key challenges and issues that area related to 5G communications and to discuss how these issues and challenges can be handled effectively in various commercial applications. In our research paper, we will attempt to emphasize some particular challenges and issues of next-generation (5G) communication as well as introduce a specific strategy to manage those challenges in this field. Moreover, a comparative analysis will be presented which will evidently make a distinction between the existing research work and our research in next-generation (5G) communication in a wireless network application.
Keywords:	Challenges, Issues, Wireless Network, Next Generation (5G), Commercial Application.
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	DESIGN ARCHITECTURE DEVELOPMENT OF IOT-BASED CATFISH CONTROL AND MONITORING
Author:	ADANI BIMASAKTI WIBISONO , RIYANTO JAYADI
Abstract:	This study aims to improve the quality and productivity of catfish farming by improving the water quality of catfish ponds. Based on survey findings, cultivators need help to check pond water pH and temperature routinely and cannot carry out continuous monitoring. In addition, water quality also needs to be monitored and maintained continuously because water quality determines 50% of the success of catfish farming in the pond. From this problem, the Internet of Things can help in continuous control, monitoring and even automation. So, this research focuses on designing an IoT system that can do these three things for catfish ponds. By using the "design thinking" method, where the approach focuses on the problems and needs of users. Various designs have been made to help catfish farmers control and monitor water quality in catfish ponds. The result of this study is a variety of designs that describe how this whole system works. The designs that have been made include activity diagrams use case diagrams, class diagrams, database architecture, user interface design, and hardware design. The system created makes it very easy for cultivators. Cultivators only need smartphones and can monitor and control from anywhere and anytime. The system that has been created can help catfish farmers improve the quality and productivity of catfish farming.
Keywords:	Catfish cultivation, Internet of Things, Aquaculture, System Design
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	THE MODERATION EFFECT OF TEACHERS' EXPERIENCE AND BIOGRAPHY ON THEIR INTENTION TO USE THE GAMIFICATION IN ONLINE LEARNING ACTIVITIES
Author:	MUTLAQ AYED ALOSAIMI, IRFAN NAUFAL UMAR, SITI NAZLEEN ABDUL RABU
Abstract:	The world is rapidly moving toward implementing different technological innovations and the Internet in learning by providing effective educational platforms. These can significantly aid teachers in helping students achieve the set learning objectives and enhance students' academic performance. In Saudi Arabia, the Ministry of Education has launched the "Future Gate Program" project as one of the national transformation initiatives toward realising digital education for several purposes. This study investigates the impact of the teachers' biography (age, gender) and experience on their intention to use gamification throughout other performance factors by implementing the 'Future Gate' application in learning. This study considers two theoretical frameworks, including UTAUT2 and TTF, to develop the proposed model in this study, which investigates the moderation effect of teachers' experience and biography on their intention to use gamification in online learning activities. The quantitative research design has examined teachers' perceptions about their intention to adopt the Future Gate platform. Moreover, using a cross-sectional statistical modelling technique, Structural Equation Modelling (SEM) is used to assess the relationship between the study's constructs. The results showed that 'age' has a significant, negative moderating effect on 'Habit and Intention to Use' (ß = - 0.365, t-value = 4.690, p-value <0.001). Likewise, the effect of 'Performance Expectancy' on 'Intention to Use' is negatively moderated by the respondents' experience (ß = -0.129, t-value = 2.165, p-value = 0.031). However, gender showed no significant moderating effect between the independent variables of the study (PE, EE, SI, HM, and H) and Intention to Use. Accordingly, teachers' intention to use showed a significant negative impact due to their age and habit. Also, previous experience with performance expectations negatively influences the intention to use.
Keywords:	Education, Technology, Distant Learning, Hedonic Motivation, Saudi Arabia, Tablet, Gamification, Structural Equation Modelling, Future-Gate Platform.
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	THE USE OF SMARTHPHONES TO BUILD CONCEPTUAL UNDERSTANDING OF ELEMENTARY SCHOOL STUDENTS
Author:	IMAS SRINANA WARDANI, ARI WIDODO, MUNIR
Abstract:	Most elementary school students have difficulty understanding the concept of the human respiratory system. This happens because their science study tends to be rote learning. To overcome this, teachers use smartphones because all students have smartphones and their smartphones have various features that can help students learn. The purpose of this research is to analyze the influence of smartphone use on elementary students' understanding of concepts in learning activities. This research uses a Quasi-experimental method. The research design used is the nonequivalent control group design type. The sample consists of 21 students who carried out their learning activities using smartphones and the other 22 students who carried out their learning activities without using smartphones. Research data was collected using a multiple-choice test consisting of 48 questions focusing on learning materials about the human respiratory system. The results of this research indicate that there are significant differences in students' scores regarding conceptual understanding tests after they learned the human respiratory system materials by using smartphones. This is because camera features available on smartphones help students to take pictures and record so that students can be directly involved in the learning activity and can observe again and again the experimental activities they carried out. Thus, it can be concluded that learning activities using smartphones on the human respiratory system can improve students' understanding of concepts. This research contributes to the quality of learning so that schools can apply smartphone use to other subject matter. In addition, future researchers can conduct similar research with different research materials and focus, for example improving communication skills.
Keywords:	Smartphones, Building Understanding Of Concepts, Elementary School Students
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	PREDICTIVE MAINTENANCE USING RNN AND LSTM MODELS
Author:	KAMPALLI RAMU, KUMAR NARAYANAN
Abstract:	The art of farming is the oldest and challenging factor in human life. In this fast phased environment and with the increase in the destruction of atmosphere and other natural resources, it is very questionable to acquire quality crops. This paper focuses to predict options which control and track the natural factors which are involved in the agriculture system. This work focuses on analyzing different features of crop and initiate predictive maintenance activities for all the sensors associated with that farm land. This ac-tivity facilitates the farmer with sensor failure reduction and helps in effective monitoring of the crop. Different factors like humidity, soil temperature and the luminosity of the crops are considered for effec-tive maintenance activity. This work is implemented using the forward and backward propagation algo-rithms using certain attributes of dataset. This paper facilitates an effective prediction ecosystem after investigating the numeric data collected from different sensors attached to plants which are meant for earlier failure prediction of those devicesdepending on the trained data. Using the forecast model and analyzing time-series data, LSTM model has obtained good accuracy with almost 97% accuracy.
Keywords:	Agriculture, PM, Natural conditions, luminosity, Machine Learning
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	FACTORS THAT AFFECT THE SUCCESS RATE OF DIGITAL TRANSFORMATION CONCERNING THE COMPANYS PERCEIVED PERFORMANCE ON INDONESIAN CONSTRUCTION STATE COMPANY
Author:	FERALIANA AUDIA UTAMI , RIYANTO JAYADI
Abstract:	Implementation of Digital Transformation lately has become a necessity. However, PT Nindya Karya (one of the Indonesian construction state companies) faced many challenges during its digital transformation process. Its Challenges are not only in terms of technological readiness but also the readiness of its employee and commitment from top-level management. This paper aims to determine what factor significantly affects the digital transformation success rate of PT Nindya Karya. The survey has been distributed using a 5-point Likert scale. The questionnaire received 256 responses from PT Nindya Karya employees. The data were analyzed using Smart-PLS version 3.3.9 and applying PLS-SEM analysis, including inner and outer model testing. The structural model achieved a good fit (SRMR = 0.055, NFI = 0.802). The research findings show that top management commitment, IT capability, work culture change, and human capital capacity significantly affect digital transformation. In addition, the research findings also show that the influence of top management commitment to digital transformation mediated by IT capability, the digital transformation itself, has a significant effect on the company's perceived performance. By knowing the factors that affect digital transformation, company stakeholders can improve the strategies related to digital transformation in their company.
Keywords:	Digital transformation, Top Management Commitment, IT Capability, Work Culture Change, Company Perceived Performance
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	INVESTIGATING THE EFFECTS OF DATA AUGMENTATION TECHNIQUES ON BRAIN TUMOR DETECTION ACCURACY
Author:	ASHRAF M. H. TAHA, DR. SYAIBA BALQISH BINTI ARIFFIN, SAMY S. ABU-NASER
Abstract:	Brain tumor detection is an essential task in medical image analysis. Convolutional neural networks (CNNs) have shown remarkable performance in various computer vision tasks, including brain tumor detection. However, the performance of CNNs depends heavily on the availability of large and diverse training data. In medical imaging, acquiring a large dataset is often challenging due to ethical and practical issues. Data augmentation is a widely used technique to overcome this limitation by generating additional training samples from the existing dataset. In this research paper, we investigate the impact of data augmentation on brain tumor detection using a deep learning approach. We compare the performance of a CNN-based model trained on augmented and non-augmented data using the BraTS 2019 dataset. The experimental results show that data augmentation improves the performance of the model significantly, achieving a higher accuracy, sensitivity, specificity, and dice coefficient in tumor detection. Our findings demonstrate that data augmentation is an effective technique for enhancing the performance of CNN-based models in medical image analysis tasks, particularly in situations where large and diverse datasets are not available.
Keywords:	Brain Tumor, datasets, deep learning, Data Augmentation
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	ENHANCING CONGESTION CONTROL AND QOS SCHEDULING USING NOVEL RATE AWARE-NEURO-FUZZY ALGORITHM IN MANET
Author:	S. MOHAN , DR. P. VIMALA
Abstract:	Mobile Ad Hoc Networks (MANET) provide a vibrant atmosphere wherein data may be substituted deprived of the necessity of human authority or a centralized server, as long as nodes work together for routing. As long as security throughout the multipath routing protocol and data transfer over many routes in a MANET is a difficult problem, this work offers a message security technique. This study presents the congestion control and QoS scheduling mechanism. The goal of this study is to examine standardized MAC protocols on MANET, to measure performance under various node densities and MAC protocols. Initially, this work presents the Centralized Congestion Detection method to detect congestion with baseline parameters. Accordingly, the congestion is avoided using Novel Rate Aware-Neuro-Fuzzy based Congestion Controlling strategy. This method effectively controls the congestion in the Network. This mechanism has been proposed which defines three levels of congestion based on which the data rate, throughput, overhead and delay. However, after controlling the congestion, the optimal routes are given to the packets by proposing an Ambient Intelligence-based Ant colony optimization quality-aware energy routing protocol (AIACOAR). This method finds the most efficient route to a destination and decreases the time and energy required. Accordingly, for securing the network against malicious attacks, an Elliptic Curve Cryptography (ECC) encryption mechanism is presented. Consequently, the multihop scheduler performs QoS-based scheduling in MANET. Schedulers in MANET take into account various QoS parameters such as end-to-end packet delay, packet delivery ratio, flow priority, etc. The proposed method is implemented using Matlab software, and the evaluation metrics are PDR, jitter, congestion detection time, delay, route selection time, and throughput. The performance of the proposed method is compared to the existing AIFSORP and LF-SSO techniques. While compared to these methods, the proposed method’s performance is improved in terms of PDR, delay, throughput, etc. The PDR value of the proposed method reaches approximately 99%, and it produces a very low delay. This produces reliable route discovery, optimized congestion control, and better QoS scheduling, therefore, these improve the system performance. In future, a recent bio-inspired technique is presented to even more minimize energy consumption and further improve the system's performance.
Keywords:	MANET, MAC Protocols, Congestion Control, Routing, QoS Scheduling, Elliptic Curve Cryptography, Ambient Intelligence, and Ant Colony Optimization.
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	SINGLE INPUT AND MULTIPLE OUTPUT DC-DC CONVERTER FOR ELECTRIC VEHICLE APPLICATIONS
Author:	HEMALATHA JAVVAJI, GUDAVALLI MADHAVI, VEMULAPALLI HARIKA, G.VEERANNA, MUZEEB KHAN PATAN, MAJAHAR HUSSAIN MAHAMMAD
Abstract:	In this paper a DC-DC single input multi output (SIMO) converter is developed for electrical vehicle applications. In the proposed converter the output port terminals can be incremented. Such multi-port converters are increasingly playing a key role in electric vehicle applications. Designing SIMO converters still faces difficulties due to the cross-regulation issue. In order to get beyond the earlier described restrictions, a SIMO topology is suggested in this work. With regard to duty cycle and inductor currents, it is capable of producing three different output voltages. Different single-input multi-output (SIMO) converter configurations are described in the literature. The majority of SIMO converters generate outputs with operating restrictions on duty ratio and inductors' charging. The proposed topology does not have cross regulation issues, hence changes in output current have no effect on the load voltage. During control, the loads are kept separate.
Keywords:	DC-DC Converters, Multiports, Duty Ratio, SIMO Converter
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text

Title:	REDUCED COMPONENTS MULTILEVEL INVERTER TOPOLOGY
Author:	K.S.S.PRASAD RAJU , K.VAISAKH
Abstract:	This paper presented a new low-component-count multilevel inverter. The generalized structure of the proposed topology is formed via cascading, and it is found to improve output voltage level count. It is possible to use the developed structure of the proposed topology with either asymmetrical or symmetrical values of DC source voltages. Binary configuration is employed to choose the values of DC source voltages. Third harmonic injection sinusoidal pulse width modulation technique is adopted to create gating signals for the presented topology. Comparative analysis of the presented topology shows the best performance in DC sources. The switch count and low blocking voltage on the switches make this topology more cost effective compared to certain other topologies in the extant literature. Losses analysis for the presented topology was shown. MATLAB/SIMULINK is employed to carry out the simulation. The presented topology is validated using a prototype designed to create a 7-level output voltage.
Keywords:	Modeling Of Switching Pulses, POD Technique, Symmetrical And Asymmetrical Multilevel Inverter, Selection Of DC Sources.
Source:	Journal of Theoretical and Applied Information Technology 15^th June 2023 -- Vol. 101. No. 11-- 2023
Full Text