		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
April 2024 | Vol. 102 No.7

Title:	HARVESTING BODY HEAT FOR REMOTE PULSE SENSOR MONITORING
Author:	MAJDA LAKHAL, MEHDI TMIMI, ABDELALI IBRIZ, MOHAMED BENSLIMANE
Abstract:	With the evolution of technologies, energy consumption management has become a paramount concern. Energy harvesting is employed to recharge, supplement, or replace batteries in systems where their use is impractical, costly, or dangerous. Ambient energy sources such as light, temperature differentials, mechanical vibrations, or RF signals can be converted into electrical energy using a transducer. These energy sources are ubiquitous and can be harnessed to power electronic devices. In this article, we focus on the pulse sensor, which plays a crucial role in monitoring heart rate and displaying real-time corresponding graphs. The objective is to remotely monitor patients' cardiac health while minimizing energy consumption. To achieve this, we explore an innovative solution that involves generating thermoelectric energy from body heat to power the back electrodes of the pulse sensor. This approach reduces dependence on external power sources and optimizes energy efficiency in the context of telemedicine. By combining ambient energy harvesting with the utilization of thermoelectric energy, we offer a promising solution to enhance remote heart rate monitoring. This groundbreaking approach allows us to leverage the benefits of telemedicine while ensuring efficient energy utilization. The article delves into the details of this technology and highlights its potential in the realm of connected health.
Keywords:	Pulse Sensor, Telemedicine, Thermoelectric Energy, Seebeck, Peltier
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	OPTIMIZED DEEP LEARNING ARCHITECTURE FOR THE EARLY-STAGE CANCER DETECTION IN BREAST IMAGES
Author:	B. SRINIVAS, DR. M. SRIRAM, DR.V. GANESAN
Abstract:	Breast cancer is characterized by the uncontrolled growth of cells in the breast tissue and is a prevalent and potentially deadly disease. This type of cancer can manifest in various forms, such as ductal carcinoma in situ (DCIS) or invasive ductal carcinoma, and it affects both women and, in rarer instances, men. The complexity of breast cancer arises from its heterogeneity, with distinct subtypes having different biological behaviours and responses to treatment. Early detection through routine screening, including mammography and clinical examinations, significantly improves prognosis and treatment outcomes. The proposed Image Augmentation Flemingo Optimization Deep Learning (AFO-DL) method introduces an innovative and comprehensive framework for breast cancer detection using medical imaging. This methodology integrates three key components: image augmentation, Flemingo Optimization, and deep learning techniques. Image augmentation diversifies the training dataset through various transformations, enhancing the robustness of deep learning models. Flemingo Optimization introduces a specific optimization strategy tailored to the complexities of breast cancer-related tasks. Leveraging artificial neural networks, deep learning methods facilitate complex pattern recognition and feature extraction, contributing to improved accuracy in breast cancer detection. The AFO-DL framework aims to provide a novel and effective approach for advancing the capabilities of deep learning models in breast cancer diagnosis, potentially leading to more accurate and reliable outcomes in medical imaging analysis. The comprehensive integration of these elements demonstrates the potential of AFO-DL as an impactful tool in the field of medical imaging for breast cancer detection.
Keywords:	Augmentation Flemingo, Breast cancer, Deep Learning, Artificial Neural Networks, Image Augmentation
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	OPTIMIZED BACKGROUND SUBTRACTION FOR HIGH-PERFORMANCE VIDEO TEXT DETECTION
Author:	SRIDHAR GUJJETI, DR. M SRIRAM, DR. V. GANESAN
Abstract:	Video text detection is a critical aspect of computer vision with applications spanning from content analysis to accessibility. The complexity of text detection in video frames arises from dynamic backgrounds, variable lighting conditions, and diverse text fonts and sizes. This paper presents a novel and efficient approach to text detection in video processing, leveraging advanced optimization techniques to enhance the performance of Background Subtraction (BS). The proposed model integrates fuzzy 2-partition entropy, Background Crunch Optimization (BGCO), and improved fuzzy C means clustering to create an innovative BS algorithm specifically tailored for text detection. The model's adaptability and simplicity make it suitable for a wide range of video processing applications. The experimental evaluation showcases the sensitivity of the proposed model to hyperparameter tuning, with optimal values for learning rates, batch sizes, and epochs significantly impacting performance. The results, detailed in optimization tables, highlight the importance of fine-tuning these parameters for achieving peak accuracy. Furthermore, the classification results demonstrate the model's robustness, consistently achieving high accuracy, recall, precision, and F1-score across multiple runs. The model exhibits a remarkable ability to accurately detect and classify text instances in video frames. The proposed approach contributes to the field of text detection in video processing, offering a comprehensive solution with practical applications. The integration of advanced optimization techniques enhances the efficiency of BS, making the model promising for real-world scenarios where accurate text detection is crucial.
Keywords:	Background Subtraction, Fuzzy C-Means, Crunch Optimization, Text Detection, Deep Learning, Video Processing
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	OPTIMIZED DEEP LEARNING MODEL FOR SENTIMENTAL ANALYSIS TO IMPROVE CONSUMER EXPERIENCE IN E-COMMERCE WEBSITES
Author:	B. RAJU, DR. M. SRIRAM, DR. V.GANESAN
Abstract:	Sentiment analysis plays a pivotal role in deciphering customer sentiments from vast amounts of unstructured data, particularly in the context of e-commerce where customer reviews are prolific. The evolution of e-commerce reviews toward a multimodal format, including images, videos, and emojis, introduces new dimensions to sentiment analysis. Traditional text-based models may struggle to effectively capture sentiments expressed through non-textual elements. This paper proposed an effective sentiment analysis model for the E-Commerce Platform to improve the user consumer experience. The proposed method comprises Fejer Kernel filtering for data points estimation in the E-commerce dataset points. Within the estimated data points fuzzy dictionary-based semantic word feature extraction is performed for the estimation of features in the E-Commerce dataset. The dataset for the analysis is computed with the Optimized Stimulated Annealing for the feature extraction and selection. The classification of customer opinion is classified with the BERT deep learning model. The feature extracted from the model is the opinion of consumers in the E-Commerce dataset. The classification of consumer preference experience is based on opinion of customers in the E-commerce dataset. Simulation results demonstrated that proposed model achieves the higher classification accuracy for the E-Commerce platform.
Keywords:	Sentimental Analysis, Deep Learning, BERT, Fejer-Kernal, Stimulated Annealing, E-Commerce
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	REVOLUTIONIZING RICE YIELD PREDICTION: A DATA-DRIVEN APPROACH IN MAURITANIA
Author:	CHEIKH ABDELKADER AHMED TELMOUD, MOHAMEDOU CHEIKH TOURAD
Abstract:	In agricultural development, accurate forecasting of crop yields is crucial to ensure food security and resource allocation. This study aims to demonstrate the power of data by optimizing machine learning models to predict increased rice yields. We use a holistic approach that combines advanced machine learning techniques, robust data prioritization, and feature engineering to extract meaningful insights from climate and agricultural data using models such as random forest regression (RFR) and gradient boosting regression (GBR). achieved remarkable accuracy in predicting grain yield. Through extensive testing and analysis, we show that our model is superior to traditional methods such as K-Nearest Neighbor (KNN), including Long-Term Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) We our findings highlight the potential of optimized machine learning models to modify rice yield forecasts Farmers and policy makers are empowered to make informed decisions with valuable insights. By harnessing the power of data, we are paving the way for sustainable agricultural practices and making important contributions to global initiatives aimed at achieving food security, especially in Africa.
Keywords:	Agriculture, IoT, KNN, LSTM, RFR, GRU
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	HYBRID TRAINING IN ROBOTICS AND THE DEVELOPMENT OF PROFESSIONAL TECHNOLOGY SKILLS FOR TEACHERS IN MOROCCO
Author:	IDRISSI LOUKILI BTISSAM, BABOUNIA AZIZ, ROUAINE ZAKARIA, NIYA HANANE
Abstract:	The hybrid training of teachers in Information and Communication Technology for Education (ICTE) gives rise to a new field of research in artificial intelligence for pedagogical purposes. With the rise of robotics, particularly in the field of education, its integration into teacher training appears necessary. This investigation aims to identify the impact of hybrid training in robotics through online institutional platforms on the professional skills development of teachers in the public primary education sector.Employing a quantitative approach to assess the impact of this robotics training on the development of teachers' skills in the public primary sector, using Generalized Linear Models (GLM), specifically the "logit" model, under the application of binary logistic regression. This analysis explicitly demonstrates that these training programs, such as humanoid robots, animal robots, robotic kits, and educational robots, have a significantly positive effect on improving teachers skills.
Keywords:	ICTE, Humanoid Robot, Robotic Kit, Animat Robot, Educational Robot, Teacher Skills Development, Binary Logistic Regression.
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	ANALYSIS OF CONTEMPORARY METHODS OF INTEGRATING EMOTIONAL INTELLIGENCE INTO ARTIFICIAL INTELLIGENCE SYSTEMS: ADVANTAGES, DISADVANTAGES, AND PERSPECTIVES
Author:	MYKHAILO ZHYLIN, VIKTORIIA MENDELO, YULIIA CHUMAIEVA, ANDREY KERNAS, YEVHEN ZAPOROZHTSEV
Abstract:	Artificial Intelligence (AI) has started to dominate the technological sector in recent years. It began by offering us a fresh perspective on developing and organizing the rapidly evolving technology trends. It provided us with several answers to the issues we confront and enhanced the effectiveness of the solutions we have already accomplished and are still overcoming. It still goes through the process of becoming flawless, though, because it can still acquire knowledge and develop into the cognitive abilities of humans. Through observation and experience, AI may be able to respond more quickly than a person, however, as of right now, this ability is limited to specific domains. Even if this intelligent agent has a great influence on society now, it might be more compassionate if it had emotional intelligence built into it. We can enable AI to expand its domains of expertise and offer more sophisticated answers to challenging issues with its use of emotional intelligence. The gap between a person and a computer may disappear if we can create Emotional Artificial Intelligence. This may assist in a variety of professions, including medicine, consulting, education, and more, and it can open new chances by treating everyone equally. The article will discuss the role that emotional intelligence plays in AI, its applicability across a range of fields, and its potential effects on society.
Keywords:	Artificial intelligence, Emotion Intelligence, Fields, Advantages, Disadvantage, Integration.
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	A HYBRID MODEL OF RSA AND NTRU FOR SECURING OF CLOUD COMPUTING
Author:	KHALID ALTARAWNEH, IBRAHIM ALTARAWNI, MOHAMMED ALMAIAH, MUSTAFA HAMMAD, TAYSEER ALKHDOUR, ROMMEL ALALI, ABDALWALI LUTFI, MAHMAOD ALRAWAD
Abstract:	Cryptographic techniques are essential in guaranteeing the security and confidentiality of data in cloud computing, which is a critical concern. In order to provide a strong and secure solution for cloud computing environments, this study presents a novel hybrid cryptographic architecture that combines the characteristics of RSA for key exchange with NTRU for data encryption. The suggested hybrid paradigm places an emphasis on effective cloud key management and the safe implementation of cryptographic operations. The approach creates a secure framework for cryptographic key distribution by using RSA for key exchange, guaranteeing the confidentiality of data while it is being transmitted. Data encryption using NTRU, a quick and effective encryption method, improves the security of data processed and stored on the cloud. Because of its strong encryption capabilities, NTRU is a good fit for the requirements of cloud computing, where security and performance are critical factors. This hybrid model's RSA and NTRU synergy provides a well-rounded solution to the security issues associated with cloud computing. With RSA's key exchange capabilities and NTRU's robust data encryption, the model offers a comprehensive solution that secures data in the cloud during its entire lifecycle. In conclusion, this work presents a secure, effective, and flexible hybrid paradigm combining NTRU and RSA cloud computing security. This approach helps enterprises safeguard their data while making use of cloud computing technologies by highlighting the critical components of secure cryptographic operations and cloud key management.
Keywords:	Cryptography, Security, RSA, NTRU, Cloud Computing.
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	IMPROVING THE PERFORMANCE OF ORTHOGONAL MULTIPLEXING FREQUENCY DIVISION USING EFFICIENT CHANNEL ESTIMATION MODEL
Author:	ASWANI LALITHA, G.HARINATHA REDDY
Abstract:	Multiple input and multiple output – Orthogonal Multiplexing Frequency Division (MIMO-OFDM) has been a trivial focus and is now a potential to use high-rate and robust data technology in wireless systems. High spectral efficiency and output over fast fading channels are difficult to achieve simultaneously. Fading is a phenomenon that results in differences in the duration of the channel intensity due to small effects of the multipath. The use of a single transmitter to that of a common receiver may cause these interferences. Orthogonal Frequency Division Multiplexing is the digital modeling technology used for multi-carrier. Unlike wireless channels, it plays a significant role in the transfer of signals in broadband wireless networking. The multi-way channel is where OFDM really shines. With OFDM, a parallel channel is created out of a selective frequency channel. The waveform's orthogonal area maintains the frequency separation of the various carrier frequencies, which utilize the sub channels. The goal of channel estimation, an optimization issue, is to find the optimal value of the channel's estimated coefficient while minimising the gap between the two. The channel allotment plays a key role in avoiding interference with node-to-node communications in sensor networks, typically involving centralised coordination. Increasing numbers of devices or terminals and the complex network environment will place a significant burden on wireless networks for computing and information interchange during the centralised delivery of the channel. The proposed model introduced a Dynamic Channel Estimation Model (DCEM) for accurate channel estimation and allocation for improving the network performance. The key goal of research is to learn how to solve these problems in order to increase the performance of a communication system by proper channel estimation and channel allocation. Compared to conventional approaches, the model proposed and its findings show that the performance of the model proposed is higher.
Keywords:	Channel Estimation, Orthogonal Multiplexing Frequency Division, Multi-Way Channel, Optimization, Fading, Collision Reduction.
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	IMPLEMENTATION AND TESTING OF A PROPOSED DOMAIN MODEL FOR ADAPTIVE HYPERMEDIA
Author:	KAMAL OKBA, MEHDI TMIMI, KAMAR OUAZZANI, MOHAMED BENSLIMANE
Abstract:	This paper joins our implemented and tested works regarding the three fundamental models of the adaptive hypermedia systems. Namely: learner model, domain model and adaptation model. In a previous work, we introduced our proposal of the domain model that describes how the knowledge and concepts of courses are structured within the adaptive hypermedia systems. Our proposal was based on new relevant ideas and structures that were designed mainly to be more suitable for adaptation purposes. And as an extension of our workflow, we have moved from the conceptual phase to the implementation phase of this model which we will discuss in detail in this article by first discussing the small changes that we made to our model at the conceptual level, and then introducing the different web tools that we developed to manipulate our domain model, and finally presenting screenshots of our tested tools that proves the efficiency and utility of our proposed model.
Keywords:	Adaptive hypermedia system, Domain model, Adaptation model, Learner model, E-learning
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	SENTIMENT ANALYSIS FOR TIKTOK SHOP'S CLOSURE IN INDONESIA USING NAIVE BAYES MODELS AND NLP
Author:	HENOCH JULI CHRISTANTO, STEVEN SONDRA ALLEN WIDODO, CHRISTINE DEWI, YERIK AFRIANTO SINGGALEN, DALIANUS RIANTAMA, ANDRI DAYARANA K. SILALAHI
Abstract:	Sentiment Analysis (SA), or opinion mining, is a task in natural language processing (NLP) that entails identifying the sentiment conveyed in a text, such as positive, negative, or neutral. Multiple methodologies and strategies exist for conducting sentiment analysis, from conventional procedures to more sophisticated machine-learning techniques. This study applies Sentiment Analysis (SA) techniques with NLP approaches to gauge sentiments related to TikTokShop’s closure in Indonesia. The study uses Twitter data to analyze sentiments using different algorithms such as the Multinomial Naive Bayes, the Bernoulli Naive Bayes, and the Complement Naive Bayes. Moreover, it utilizes a Count Vectorizer and TF-IDF Vectorizer to enhance sentiment analysis. Furthermore, using TextBlob with the CountVectorizer approach is the most accurate at 86.60% in sentiment classification. The analysis sheds light on sentiment analysis techniques applicable to TikTokShop closure as well as which algorithm and vectorization approach can be used to measure sentiments derived from the Twitter data.
Keywords:	Sentiment Analysis, TikTokShop Closure, Twitter Data, TextBlob
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	DESIGN AND EVOLUTION OF MAC ALGORITHM BASED STRATEGY TO MITIGATE BLACK HOLE ATTACKS IN WIRELESS SENSOR NETWORK
Author:	SUBHRA PROSUN PAUL, D. VETRITHANGAM, G. KRISHNA MOHAN3, THIYAGU. T, SUBRAMANIAN SELVAKUMAR, M. MAITHILI SAISREE, NIMMAGADDA CHANDRA SEKHAR
Abstract:	Despite tremendous advances in successful packet transmission in the current context of technological progress on wireless networks, network security remains an unavoidable issue in this field due to various wireless network attacks. A black hole attack is one of the most crucial threats to wireless network security. In a black hole attack, a malevolent node announces openly about the availability of the shortest route throughout the wireless network, which is totally false. Whenever a participating node forwards its packet to that malevolent node, the packets are dropped. In order to provide high-level security during packet transmission throughout the sensor network, a strong threat handling mechanism is required. In this research paper, the problem statement is to introduce the MAC algorithm-based black hole attack avoidance mechanisms, where a shareable secret key concept is used during the packet transmission process throughout the network. The proposed technique's algorithm, as well as the mathematical explanation and how the MAC algorithm is implemented, are thoroughly discussed in this paper. Using MATLAB software, the proposed algorithm is simulated, the shortest paths are identified, and the shortest path distance and time are calculated as simulation results. Furthermore, by providing a comparative analysis in this article, we have attempted to identify the clear differences between our proposed mechanisms and the existing techniques in this field.
Keywords:	Black Hole Attack, MAC Algorithm, Wireless Sensor Network, Design, Detection, Security, Routing System
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	DESIGN OF RECOMMENDENDATION SYSTEMS USING DEEP REINFORCEMENT LEARNING – RECENT ADVANCEMENTS AND APPLICATIONS
Author:	KRISHNAMOORTHI.S, GOPAL K. SHYAM
Abstract:	The paradigm of recommendation systems (RS) has witnessed remarkable evolution in terms of providing accurate recommendations to the users. However, it is a complex task to generate appropriate recommendations to the users. In this context, RS use Artificial intelligence (AI) based techniques to recommend products based on the customer’s preference. The adaptability of these techniques suffer from complexities systems such as data availability, changes in the user preferences, and unpredictable items. This motivates the researchers to emphasize performance enhancement of RS by overcoming these problems. This review focuses on the implementation of deep reinforcement learning (DRL) algorithms for RS. The study discusses different design aspects of RS and summarizes DRL-based techniques applied for recommendation systems. In addition, this review analyzes the challenges and relevant solutions based on the existing literary works. This paper also discusses the open issues of DRL and highlights the potential research directions in the RS field.
Keywords:	Comparative Analysis, Deep Reinforcement Learning, Policy Optimization Algorithms, Recommendation Systems
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	BREAST CANCER DETECTION USING DEEP LEARNING ON BIOMEDICAL MAMMOGRAM IMAGES
Author:	PRETOM ROY CHOWDHURY, MOHAMED EL-DOSUKY, SHERIF KAMEL
Abstract:	Millions of women worldwide are affected by breast cancer, which is a serious global health issue. The likelihood of successful therapy and the prognosis both greatly benefit from early identification. The most popular screening method for breast cancer, mammography, produces precise biological images that can help with the early detection of malignancies. However, it is still difficult to correctly interpret mammography pictures, which frequently results in false positives or negatives. This study attempts to create a biological mammogram based deep learning system for breast cancer diagnosis. Convolutional neural networks (CNNs) are used to automatically identify and analyse mammogram pictures in the proposed system, enabling radiologists to make quicker and more accurate diagnoses. To ensure the best performance during the training phase, these photos underwent preprocessing to reduce noise and enhance characteristics. The deep learning model used is a cutting-edge CNN architecture that was pretrained on a sizable dataset to fully utilise its learned representations. The deep learning model underwent thorough training, validation, and fine-tuning procedures to ensure robustness and generalizability. A variety of data augmentation methods, including rotation, scaling, and flipping, was used to enlarge and diversify the dataset during training. To further increase the model's accuracy, transfer learning was used to utilize knowledge from other similar tasks. Using a variety of criteria, such as sensitivity, specificity, accuracy, and F1 score, and the performance of the created breast cancer detection system was carefully assessed. The results showed a substantial increase in accuracy when compared to traditional mammography analysis methods. The method demonstrated impressive specificity in reducing false positives and sensitivity in identifying actual positive situations.
Keywords:	Convolutional Neural Network Hybrid Architecture, Deep Learning, Transfer Learning
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	REVOLUTIONIZING POTATO LATE BLIGHT SURVEILLANCE: UAV-DRIVEN OBJECT DETECTION INNOVATIONS
Author:	YASSINE ZARROUK, MIMOUN YANDOUZI, MOUNIR GRARI, MOHAMMED BOURHALEB, MOHAMMED RAHMOUNE, KHALID HACHAMI
Abstract:	The ongoing integration of cutting-edge technologies is profoundly transforming agricultural oversight, where drones emerge as pivotal instruments for precise crop monitoring, early disease detection, and efficient land management. The harmonious synergy between drones and AI, specifically deep learning, is revolutionizing the surveillance of plant diseases, facilitating accurate realtime detection. This innovative approach not only promises enhanced effectiveness but also fosters sustainable agricultural management, steering the course of modern farming towards intelligent and environmentally conscious practices. This article undertakes a thorough comparative exploration of recent advancements in deep learning-based object detection. It investigates two model families - the single-pass YOLO (You Only Look Once) and the two-pass RCNN (Region-based Convolutional Neural Network) - along with their respective variations, with a particular focus on their potential use in drone-based agricultural surveillance, specifically targeting the detection of Potato Late Blight. The conducted experiments unveil promising results across various metrics, affirming the invaluable role of this tool in the detection and monitoring of agricultural diseases. This research not only contributes to advancing our understanding of deep learning in agricultural contexts but also underscores the significance of integrating cutting-edge technologies for sustainable and efficient farming practices.
Keywords:	Potato Late Blight, Deep Learning, Computer Vision, Drones, UAV, Object Detection
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	AN ALGORITHM FOR GENERATING DESCRIPTIVE SENTENCES OF THE HUMAN HEAD PARTS BASED ON ENGLISH GRAMMAR
Author:	LOAI ALAMRO, YUHANIS YUSOF, NOORAINI YUSOFF
Abstract:	Human head exhibits many biological features (attributes) that represent the characteristics of the human head with robust inherent stability and individual variation. These attributes provide important discriminative knowledge about humans, such as gender, age, race, hairstyle, hair color, etc. Recently, several human head attribute classification networks have been proposed. However, these networks do not provide a clear picture of the human head because they predict head attributes in terms of binary values (i.e., 0 or 1) or by their labels (i.e., male, young). Therefore, in this study, a description algorithm was proposed to describe the main characteristics of the human head using the adjective’s arrangement rules. The proposed algorithm was reviewed by experts, and the responses of seven experts show that the algorithm follows the adjective’s arrangement rules in accordance with the conventions of human language. The experts also found the descriptive sentences acceptable, understandable, and grammatically correct.
Keywords:	Human Attribute Classification, Human Attribute Description, Object Identification, Object Recognition, Deep Learning.
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	DIGITAL CURRENCY OF THE CENTRAL BANKS: TRENDS OF THE EURO AREA AND PROSPECTS OF THE USE WITHIN THE IMPLEMENTATION OF THE EUROPEAN GREEN DEAL
Author:	OKSANA HRUBLIAK, OLHA POPELO, KOSTIANTYN SHAPOSHNYKOV, ARTUR ZHAVORONOK, NATALIA OSTROVSKA, DENYS KRYLOV
Abstract:	In the article, using the certain key financial concepts that are proposed in the scientific literature, the current experience and prospects of the digital currency of the central banks (CBDC) use in the global financial market are investigated. The study of the possibilities of the CBDC introduction in the world has showed that only 14 countries of the European Union of the 27 are at the development stage, the rest have not started the work on the introduction of this financial instrument. In comparison with the global trends, this is a negative indicator. The stages of the implementation of the wholesale, retail and hybrid CBDC are characterized. In the article, the possibilities of the CBDC use are proposed, the problem points that the country may face as a result of the wide use of this financial instrument are presented. The practical significance of the authors’ conclusions and recommendations lies in the fact that a clear idea of the current barriers and prospects of the CBDC introduction in the European financial market, the possibility of its use in the transactions between banks and financial intermediaries, in the implementation of big investment projects, in particular, within the European Green Deal has been created.
Keywords:	Digital Currency, Central Banks, Global Financial Market, European Green Deal, Fintech, Green Financing, Green Technologies, Sustainable Development, Financial System, Financial Instruments, Financial Intermediaries
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	INTEGRATING INFORMATION TECHNOLOGY IN ENGINEERING GEOLOGY: CASE STUDIES AND EDUCATIONAL APPROACHES FOR ENHANCED GEO-LOGICAL ANALYSIS FOR INFRASTRUCTURE DEVELOPMENT
Author:	MALABIKA ADAK, ALTAF USMANI, ANIRBAN MANDAL
Abstract:	The paper critically investigates the indispensable geotechnical knowledge crucial for effective engineering practice and proposes innovative approaches to seamlessly incorporate it into the education and training of geotechnical engineers. It initiates by scrutinizing the core responsibilities of geotechnical engineers, encompassing exploration, analysis and design, management, and construction, subsequently delving into the principal reservoirs of this knowledge, such as engineering sciences, models, software, codes of practice, judgment, and heuristics, and their pragmatic applications. Furthermore, it anticipates and discusses forthcoming trends poised to impact the profession in the foreseeable future. The paper concludes with a strong emphasis on rectifying the current disjunction between academic knowledge and practical application in the field, underscoring the paramount importance of augmenting the comprehension of geotechnical knowledge within engineering practice. It advocates for the integration of this understanding into the education and training of geotechnical engineers, fostering mutual benefits for both academia and practitioners.
Keywords:	Decision Tree, Risk Management, Geological Uncertainties, Tunneling.
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	SIMULATING BREAST CANCER TREATMENT EFFICACY: A COMPUTATIONAL APPROACH TO OPTIMIZING PATIENT CARE
Author:	MEROUANE ERTEL, AZIZ MENGAD, SAMIRA FADILI, YOUNES BOUFERMA4 OUSSAMA RHARIB, MERYEM CHAKKOUCH, SAID AMALI
Abstract:	Treatment of breast cancer with chemotherapy is common, but its effectiveness can vary significantly depending on the individual characteristics of the patient and the type of cancer. In this context, computer simulation based on machine learning can constitute a solution to optimize the treatment strategy of patients suffering from this disease. This study uses a dataset of 490 breast cancer patients, to feed a machine learning model and uses simulation techniques to simulate different treatment strategies. Machine learning algorithms, such as Naive Bayes (NB), Decision Tree (DT), Random Forest (RF), and artificial neural networks (ANN), have been evaluated for their performance. The results indicate that the RF algorithm achieved the highest accuracy rate of 76.9%, while the NB algorithm recorded the lowest accuracy rate of 66.5%.The study demonstrates that machine learning-based computer simulation can help identify breast cancer patients at high risk of metastatic relapse and predict an individualized therapeutic combination to reduce this risk.
Keywords:	Computer Simulation, Machine Learning; Modeling, Personalized Medicine; Combination Therapy; Prediction of Therapeutic Response; Breast Cancer
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	MACHINE LEARNING MODEL OF CUSTOMER BEHAVIOR ON E-BANKING TRANSACTION USING CLASSIFICATION TECHNIQUE
Author:	SEPTIAN EKA ADY BUANANTA, LAY CHRISTIAN, PARADISE
Abstract:	E-banking services provide easy access for users to make purchase and payment transactions anytime and anywhere via devices connected to the internet. However, e-banking adoption rates may vary between groups and evolve as awareness and trust in the service increases. Customers perceive products and services differently based on their experiences, beliefs, and values. Their attitudes toward a brand or product influence their decision to buy or continue using it. Therefore, Machine Learning Development can be used by businesses to analyze customer behavior through market research, surveys, and data analytics to gain insights that can inform product development, marketing strategies, and customer relationship management efforts. Knowing customer behavior allows banks to provide a more personalized user experience. This may include presenting product recommendations or providing notifications that match customer preferences. The classification results of the best-selected technique, with an accuracy rate of 98.61% and an execution time of 10 seconds, can be used as a reference to determine customer behavior in important e-banking transactions because they significantly impact business strategy and development of digital banking services. Customer behavior data is a valuable source of information that can be used to make strategic decisions. For example, some services are often used, namely bill payment and QRIS. So, the Bank can use data analysis to design more effective marketing campaigns, optimize operational processes, and design new products that suit customer needs.
Keywords:	Machine Learning, Customer Behavior, E-banking Transaction, Classification Technique, Decision Tree
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	MACHINE LEARNING TECHNIQUES FOR CYBER SECURITY
Author:	SOUMIK SUR, MOHAMED EL-DOSUKY, SHERIF KAMEL
Abstract:	Machine learning (ML) is a branch of artificial intelligence (AI) that focuses on developing computer programs that can recognize patterns in historical data, learn from it, and make logical judgements with little to no human input. Protecting digital systems, such as computers, servers, mobile devices, networks, and related data against hostile assaults is known as cyber security. Two key components of combining cyber security with ML are accounting for cyber security where machine learning is used and using machine learning to enable cyber security. This coming together may benefit us in a number of ways, including by enhancing the security of machine learning models, enhancing the effectiveness of cyber security techniques, and supporting the efficient detection of zero day attacks with minimal human interaction. The cyber security landscape has grown more complicated due to the quick development of technology, creating a number of difficulties for protecting sensitive data and important infrastructures. This project's objective is to implement three different systems using machine learning in cyber security. The first system investigates how reinforcement learning may be used to improve cyber security measures. Reinforcement learning algorithms are taught to make the best choices based on their interactions with the environment through trial and error, which can be useful in adjusting to changing cyberthreats. The second approach focuses on malware identification since evasive and polymorphic malware have proven difficult to identify using standard signature-based methods. Several machine learning and deep learning approaches are used in this effort to accurately identify and categorize dangerous software. The third solution uses machine learning and deep learning techniques to address the crucial problem of network intrusion detection. The performance of each system's machine learning models will be evaluated throughout the project using a variety of datasets alongside evaluation measures.
Keywords:	Machine learning, Cyber security, Deep learning, Network, Attack
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	ANALYSIS OF DEPLOYMENT OF SCALABLE SERVICE USING KUBERNETES IN CLOUD ENVIRONMENT
Author:	GARIBALDY HUMPHREY WATULINGAS, YANTO SETIAWAN
Abstract:	Cloud computing technology has developed very rapidly, and it is no longer exclusively being used by large companies but also by the individual level that everyone can this technology relatively easily. Some of the main advantages that drive the growth in the application and penetration of cloud computing technology in the market are cost efficiency, flexibility, reliability, and scalability. With the implementation of scalability, users can efficiently utilize the computing resources provided by cloud service providers to obtain maximum performance at minimal cost. This research will analyze the case where cloud computing technology can be used to apply a scalable service using Kubernetes running in a cloud environment and provide evidence for the performance improvement that can be achieved from implementing scalability.
Keywords:	Cloud Computing, Container Orchestration, Scalability
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	CRITICAL SUCCESS FACTOR ANALYSIS FOR CLOUD-BASED ERP PROJECTS IN INDONESIA
Author:	WENNY RESTI PRATIWI, SFENRIANTO SFENRIANTO
Abstract:	Currently, ERP systems are experiencing rapid expansion through the utilization of cloud technology. This technology holds great promise as it is expected to significantly enhance productivity, facilitate seamless collaboration among teams, and ensure robust data security and transparency. Indonesia is one of the potential cloud markets as it is supported by the government’s digitalization drive. This creates opportunities for the widespread implementation of cloud-based IT projects in Indonesia, including those involving cloud-based ERP systems. One aspect requiring careful consideration is the possibility of project failure. Certain pivotal factors warrant consideration due to their potential influence on the project's success. This article aims to identify the critical success factors (CSFs) for projects utilizing cloud-based ERP systems, with a particular focus on SAP deployed in the cloud within Indonesian projects. The research was a quantitative study that involved 138 implementors who have engaged SAP on cloud projects in the last 5 years in Indonesia. Data was analyzed by using the PLS-SEM method. 2 of the 8 hypotheses from 3 CSF dimensions were supported by the PLS-SEM results. This study provided evidence that several critical success factors related to the Organizational dimension and Technology dimension significantly and positively impacted the success of cloud-based projects, especially those related to SAP on the cloud in Indonesia. On the contrary, none of the Critical Resource Factors (CRFs) addressed the People dimension, which has been shown to significantly contribute to project success. These findings can serve as a foundation for crafting a project strategy aimed at enhancing the success potential of cloud-based ERP projects, whether they are in the planning phase or already in execution.
Keywords:	Critical Success Factors, Cloud-based ERP, SAP, PLS-SEM, IT Project Management
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	COGNITION OF ROAD TRAFFIC SIGNS: A SYSTEMATIC LITERATURE REVIEW
Author:	WANG PEIZHI, RAIHANI MOHAMED, NORWATI MUSTAPHA, NORDAYU MANSHOR
Abstract:	As a research hotspot in computer vision, traffic sign recognition has made remarkable progress in the past few years. This study provides a systematic review of the field of traffic sign recognition. Thirty-nine papers relevant to this study were manually selected for exploration from four well-known databases (IEEE Xplore, ScienceDirect, Scopus, and Google Scholar). Five questions were proposed to describe general trends in traffic sign recognition. These questions are answered by the literature review. Specifically, first, determine the literature review method and select the papers to be analyzed. Next, various algorithms and commonly used data sets for traffic sign recognition are analyzed in detail. Then, the advantages and disadvantages of various algorithms are compared, and the challenges faced in traffic sign recognition are discussed in depth. Finally, the application fields of traffic sign recognition were deeply explored. This review helps provide guidance and comprehensive information to researchers in this field.
Keywords:	Traffic sign recognition; Deep learning; Classification; Localization; Datasets
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	THE ANALYSIS OF ASSOCIATION BETWEEN VICTIM IDENTITIES AND EXECUTION FROM GENOCIDE DATA SET USING QUANTITATIVE ANALYSIS AND OCR TECHNIQUE
Author:	TODSANAI CHUMWATANA, PUNNITA SAMBATH, SOPHEAKNINE CHIM
Abstract:	Over centuries, there have been so many conflicts happening around the world both domestic and international such as war crises, separatism, or genocide. The causes of conflicts are diverse, and rooted in historical grievances, geopolitical ambitions, economic disparities, and cultural differences. These conflicts ravage countries in various ways: economic growth, infrastructure, food system, and depopulation. Genocide is one of the conflicts mainly caused by racial differences which impact a country's development as a decrease in the population. This is because human capital is a key driver of economic growth as it directly influences the productivity of the workforce. This study aims to analyze the relationship between victim identities (gender, age, occupation, nationality, education) and their execution in prison. This research utilizes data from the genocide biographic database, developed by Yale University, and the documentation center in which documents have been stored in the format of hard copies, scanned documents, images, and PDFs which need to be transformed into digital format for future analysis. The techniques used in this proposed research are web scraping and Optical Character Recognition, also called OCR, for the extraction process. For the analysis process, the research employs statistical models like the Chi-Square Test and Logistic Regression. And also, reveals data insight by using data visualization techniques to enhance the presentation of findings. The experimental studies showed that most of the victims are male and the majority of the victims are students, military, and higher education people. This might highlight the regime’s effort to change the social fabric, and also downgrade occupational status, especially among students and intellectuals by forcing them to change their status from upper class to lower class, to avoid the difficulty of controlling these people. As a result, this significant finding showed that this problem leads to capacity reduction for country development because higher education and the young generation have been regarded as valuable and potential human resources for country development in quality and quantity. This is the reason to encourage people over the world to realize on stopping conflict, which causes the world has take a turn for the worse.
Keywords:	Web scraping, Optical Character Recognition (OCR), Genocide studies, Data visualization, Data analytics
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	EXAMINING CLOUD SECURITY: IDENTIFYING RISKS AND THE IMPLEMENTED MITIGATION STRATEGIES
Author:	EMAN AL-QTIEMAT, ZEYAD AL-ODAT
Abstract:	Cloud computing is critical for modern enterprises, because of its scalability, cost-effectiveness, and adaptability. However, these benefits are offset by security risks and obstacles, which makes cloud computing a two-edged sword. Although technology offers many benefits and services, it also puts data and businesses at risk if security threats are not carefully addressed or if security flaws are discovered after the fact. This paper examines the primary security flaws caused by cloud computing and the existing countermeasures for those flaws. The paper examines cloud models, service models, and distinguishes several deployment models. An analysis of security countermeasures for cyber threats is presented. Furthermore, the work investigates the regulatory environment controlling cloud security, considering compliance frameworks and standards that influence the creation and implementation of security solutions. The study contributes to a more nuanced knowledge of the problems and opportunities in cloud security, as well as practical advice for enterprises looking to improve their cloud security posture. In addition, some recommendations for identifying and averting cloud security risks are presented.
Keywords:	Cloud, Cloud Risks, Security, Cyber Threats, Security Countermeasures
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	PREDICTIVE STUDY OF FIRE RISK IN BUILDING USING BAYESIAN NETWORKS
Author:	SANAE KHALI ISSA, HAJIRA BAKKALI, ABDELLAH AZMANI, BENAISSA AMAMI
Abstract:	Morocco, like many countries, has experienced significant development in the construction sector. Recently, various types of buildings, including residential buildings, public structures, high-rise buildings, and workplaces, have been constructed. Unfortunately, this development is accompanied by a significant increase in domestic risks. Some of these risks are associated with natural disasters such as floods, droughts, earthquakes, etc., while others result from human activities and errors like fires, gas leaks, electrical hazards, etc. The consequences include human losses, physical injuries, psychological traumas, and material damage, leading to substantial financial losses. In this paper, we focus on the study of fire risk in buildings. We present a predictive study of fire risk in buildings using the Bayesian network method. The primary focus of the study is to calculate the probability of fire ignition in buildings, which can be triggered by various factors such as poor electrical installation, gas leaks, or the presence of flammable products. Additionally, the study considers human ignorance, inadvertence, or criminal acts as potential contributors to the fire risk. The result obtained from this study identifies electrical problems, often linked to poorly maintain electrical installations or the use of degradable electrical equipment, as a potential source of the most fire ignition in buildings.
Keywords:	Fire Risk, Decision System, Bayesian Network, Fire Ignition Probability.
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	CONTROL OF AUTONOMOUS MOTORCYCLES BY MEANS OF TRAJECTORY TRACKING AND BALANCE STABILIZATION
Author:	ERIC STIVEN URIZA CACHAYA, CAMILO ANDRÉS PASCUAS PERDOMO, DIEGO MAURICIO ECHEVERRY SUAZA, JORGE BERNARDO RAMÍREZ ZARTA, RUTHBER RODRÍGUEZ SERREZUELA
Abstract:	An autonomous motorcycle is a two-wheeled vehicle that can move without human intervention. It uses a combination of sensors, cameras, and algorithms to detect its environment and make decisions on how to move. This paper shows the comparison between the control of an autonomous motorcycle using fuzzy logic and an LQR counter controller built in Matlab and implemented in an embedded system with a microcontroller. The trajectory tracking and balance stabilization of a prototype built for this purpose is performed. It is determined how the LQR control has a good behavior in front of the fuzzy logic control in front of the impulse response represented in changes of angle in the stabilization.
Keywords:	Autonomous Motorcycles, Trajectory Tracking, Balance Stabilization, Fuzzy Logic, LQR
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	A LAYERED APPROACH OF MACHINE TRANSLATION USING TRANSLATION MEMORY ON EDGE COMPUTING
Author:	SAKSHI DHOMNE, Dr. MANOJ B. CHANDAK, Dr. ABHIJEET RAIPURKAR, Dr. SUNITA RAWAT
Abstract:	This research paper delves into the limitations of traditional centralized machine translation systems and proposes a new approach that leverages edge computing. By redistributing translation tasks and cache memory operations to local edge devices, this innovative paradigm aims to mitigate the drawbacks typically associated with centralized systems. Decentralizing translation processes allows for the execution of translation and cache memory operations directly on edge devices, such as smartphones or smart speakers, eliminating the need for reliance on distant servers. This not only minimizes latency but also enhances the overall efficiency and responsiveness of machine translation services. Looking ahead, the role of edge computing in machine translation is expected to continue to grow. While cloud-based approaches are acknowledged as alternatives, the focus is on the unique advantages of edge computing. By bringing translation tasks closer to end-users and utilizing the power of edge devices, the future of translation services is set to become seamlessly integrated into everyday interactions. This paper highlights the significance of edge computing in transforming the landscape of machine translation, offering a glimpse into a future where translation processes are more accessible, efficient, and tailored to the needs of users.
Keywords:	Centralized Machine Translation Systems, Edge Computing, Cache Memory Operations, Edge Devices.
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	ENGINEERING DESIGN PROCESS WITH CLOUD BASED LEARNING MANAGEMENT TO ENHANCE INNOVATION SKILLS AND CREATIVE PRODUCT
Author:	SATHAPORN YOOSOMBOON1, SUNTI SOPAPRADIT, THANYATORN AMORNKITPINYO , PIMPRAPA AMORNKITPINYO4
Abstract:	The objectives of this research were to 1) develop an engineering design process with cloud-based learning management to enhance innovation skills and creative product and 2) investigate the academic results using an engineering design process with cloud-based learning management to enhance innovation skills and creative product. The sample were 52 Southeast Bangkok College students majoring in computer technology, who registered for an embedded system and application subject and divided into two groups: a control group and an experimental group. The research results found 1) an engineering design process with cloud-based learning management to enhance innovation skills and creative product had eight steps : 1. Define the Problem, 2. Gather Information, 3. Generate a Solution, 4. Analyse the Solution, 5. Select a Solution, 6. Implement Solution, 7. Evaluation and, 8. Reflection and 2) Post study, the experimental group had higher scores innovation skills and creative product than the control group and the criteria.
Keywords:	Engineering Design Process, Cloud Based Learning, Innovation Skills, Creative Product
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	OPTIMAL DISTRIBUTED GENERATION FOR LOSS MINIMIZATION USING SAND CAT SWARM OPTIMIZATION
Author:	A. A. S. M. ADNAN, M. H. HUSSAIN, S. R. A. RAHIM, A. AZMI, I. MUSIRIN3, M. N. K. H. ROHANI, N. HASHIM, J. A. RADZIYAN
Abstract:	Integration of Distributed Generation (DG) into the transmission system is the current paradigm for creating unique transmission grids. Grid line loss and voltage quality may suffer from unreasonably configured DG. The aim of this paper is to rationally allocate distributed generators (DGs) in the transmission network to reduce power losses and guarantee a safe and reliable power supply to the loads. The works suggests an optimal distributed generation using Sand Cat Swarm Optimization (SCSO) for loss minimization to reduce power loss while enhancing voltage stability. The proposed algorithm was simulated and evaluated using the Matrices Laboratory (MATLAB) script programming language and has been implemented on IEEE 14-bus transmission system. The results exhibit that the SCSO method is able to determine the optimal DG size and reducing total losses by 40.77 percent for DG type 1 as compared with Particle Swarm Optimization (PSO) algorithm, 38.98% at bus 10. It can be revealed that SCSO can be used by power system planners to choose the best sizing and location.
Keywords:	Distributed Generation, Sand Cat Swarm Optimization, DG Sizing, Transmission System
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	SMART CROP PREDICTICTION USING STATISTICAL TECHNIQUES OF MACHINE LEARNING THROUGH IoT
Author:	PRABHAT KUMAR SAHU, SANGAM MALLA, MITRABINDA KHUNTIA, SMITA RATH, MANJUSHREE NAYAK
Abstract:	The paper highlights the significance of agronomy in developing countries and the challenges faced in traditional farming methods, which rely heavily on human intervention. The solution proposed involves leveraging automation in agriculture, specifically using Internet of Things (IoT) sensors. By employing regular sensing and examination of crops through IoT sensors, combined with Machine Learning and statistical techniques, the system aims to predict the appropriate crop for an area that depends on factors like moisture in the soil, temperature, and humidity. The research emphasizes the need for such technology, particularly in countries like India, where agriculture is a predominant occupation. The system is designed to address issues such as repeated cultivation of the same crops and indiscriminate fertilizer use, which negatively impact crop yield and soil health. Ultimately, the proposed system intends to offer farmers insights into the best-suited crops for their land, along with information on required fertilizers and seeds, aiming to enhance profitability, encourage crop diversification, and mitigate soil pollution.
Keywords:	Smart Agronomy, Soil Fertility, Machine Learning, Statistical Techniques, Smart Crop Prediction.
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	THE INFLUENCE OF E-COMMERCE USER EXPERIENCE ON USER SATISFACTION
Author:	ARDELLE KHALIDA RIANSYAH, CELLINE VASHTI, TANTY OKTAVIA
Abstract:	User Experience plays an important role in the platforms of various industries, including that of e- commerce. User Experience helps indicate how strong overall users feel towards a given platform’s features, and whether or not they provide benefits to the user, be it facilitating user goals, providing quality services, or clearly communicating important information to users. This study aims to delve deeper into the relationship that User Experience might have with User Satisfaction, specifically among that of e-commerce application users. This relationship is investigated by testing efficiency, consistency, and communication functions of e-commerce platforms according to the answers of respondents. Upon being analyzed with the multiple regression test, findings indicate that proper user experience aspects, when combined, provide significant positive impact to users, with efficiency having the most significant impact on user satisfaction. The contribution of this study is to determine what factors in User Experience influence the overall satisfaction of users, specifically those of Gen-Z that are currently residing in the Jabodetabek area. This paper can be used by developers, researchers, and designers alike to find out which features in User Experience improve user satisfaction.
Keywords:	Satisfaction, User, Experience, E-Commerce, Gen-Z, Indonesia
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	CROP YIELDING RATE PREDICTION AND ANALYSIS USING DEEP MACHINE LEARNING ALGORITHMS
Author:	A. GEETHA DEVI, T.V.K.P. PRASAD, K.VIDYA SAGAR, M. PREMA KUMAR, BALA PRASANTHI PAVULURI, LAKSHMI RAMANI BURRA, VEERA VASANTH RAO.B.
Abstract:	The world population rate is increasing. Production of food grains is alarming with rapid change of environmental conditions. Soil parameters are also much influencing on the crop yielding rate. Precision agriculture has much significance in meeting the demands. This paper aimed to predict the yielding rate by considering the crop decease, soil parameters considered are Nitrogen, potassium, phosphorous and environmental conditions considered are ambient temperature, PH and humidity. The decease of the crop is analyzed using machine learning algorithms. ENET regression algorithm, LASSO Regression algorithm, Kernel Edge Algorithm and staking algorithm have been considered. The trained data sets are applied to various machine learning algorithms and estimated the infected level of the leaf and eventually the yielding rate of the crop. The results achieved with ENET regression algorithm, LASSO Regression algorithm, Kernel Redge Algorithm and staking algorithm are compared to interpret the bet fit algorithm for agricultural applications. Mean square error value is considered for comparison. ANET and LASSO algorithms considered maximum pixel values and neglected minimum pixels while predicting the yield rate and decease level over other algorithms. The results obtained with the above approaches are reliable.
Keywords:	Crop Yield rate, ENET, LASSO, Kernel Redge, Stacking, MSE
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	OVERVIEW OF CYBERSECURITY RISK ASSESSMENT FOR MEDICAL INFORMATION SYSTEMS
Author:	TAYSEER ALKHDOUR, MOHAMMED ALMAIAH, KHALIL IBRAHIM ALMUWAIL, MOHMOOD A. AL-SHAREEDA, THEYAZAN ALDAHYANI, RANA ALRAWASHDEH
Abstract:	In recent years, there has been a significant increase in demand for hospital information systems in healthcare institutions. Data security, on the other hand, is a significant concern with regard to using health information systems. The purpose of this research is to examine the security risk assessment of medical information systems. This study involves a systematic evaluation of the literature to provide a complete overview of previous articles and research on Security Risk Assessment in Medical Information Systems. For this research, a qualitative and descriptive research design was applied. Scientific literature, as well as recent articles from popular publications, will be evaluated and analyzed in depth in accordance with the study design. A review of the literature enables a thorough comprehension and knowledge of this subject, Security Risk Assessment in Medical Information Systems. It provides the background for the research and provides an overview of the study's relationship to a large field of study. The main objective of this research has to analyze and discuss the findings of the literature review, and to evaluate the risks and challenges. Moreover, each study was examined in terms of methodology, threats addressed, and suggested mitigations. Additionally, the study discussed the systematic review's gaps and major neglected concerns, as well as future directions in risk assessment in medical information systems.
Keywords:	Medical Information Systems; Cybersecurity; Risk Assessment; Information Security.
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	COMPARATIVE STUDY ON THE PERFORMANCE OF NHPP-BASED SOFTWARE DEVELOPMENT COST MODEL APPLYING EXPONENTIAL-TYPE LIFE DISTRIBUTION
Author:	HYO JEONG BAE
Abstract:	In this study, an exponential-type life distribution suitable for reliability analysis of system failure occurrence phenomenon was applied to the NHPP-based software development cost model, and then the performance properties were newly analyzed using the failure time data requested by the developer. For this purpose, the parameter calculations were solved by applying maximum likelihood estimation. In conclusion, first, in the attribute analysis using the function m(t), which has an important influence on the subject of this work, all models showed an attributes of overestimating the true value. But the Burr-Hatke-Exponential model was efficient by showing the smallest error. Second, in the analysis of the reference values for efficient model selection, the proposed models were found to be appropriate as they all showed a performance of over 80%. Third, as a result of exploring performance attributes (m(t), MSE and R^2, cost, release time), the Lindley model, which showed the lowest cost and fastest release time, was confirmed to be the most efficient. Through this study, the performance properties were newly explored, and the related results can be utilized as basic design data to analyze the cost attributes needed by developers in the early stages.
Keywords:	Burr-Hatke-exponential, Exponential-basic, Exponential-type, Lindley, Rayleigh, Software Development Cost.
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	REVITALIZING SCIENCE EDUCATION: UNVEILING THE POTENTIAL OF 2D ANIMATIONS TO ENHANCE UNDERSTANDING AND ENGAGEMENT
Author:	SARNI SUHAILA RAHIM, AINI AMERA ABDULLAH, SHAHRIL PARUMO, SURIATI KHARTINI JALI
Abstract:	This research paper explores the transformative potential of 2D animations in revitalizing science education by enhancing understanding and engagement among learners. A comprehensive literature review examines previous studies on educational animations, emphasizing the cognitive impacts of visual aids on learning. The article presents an evaluation of 2D Animation for Science Secondary School Learning: Acid and Alkali. The developed application was produced to assist science students to understand the contents of the topic Acid and Alkali. An experiment was carried out to assess the effectiveness of the application as a learning tool for science students. There was a total of 32 respondents, including subject matter experts, multimedia experts and students. The findings of the current study showcased the positive impact on student engagement and learning outcomes. The results may assist in advocating for a paradigm shift in science education. The research underscores the need for innovative approaches that leverage the power of 2D animations to inspire and educate learners, thereby contributing to the ongoing revitalization of science education.
Keywords:	Animation, Science, Teaching and Learning, Multimedia, Education
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	CROSS CENTROID MERKLE SEARCH WITH LINGUAL-MULTI KEYWORDS OVER ENCRYPTED CLOUD DATA
Author:	DIVYA SURENDRAN, DR. K. SASIKALA
Abstract:	One of the most significant services provided by cloud computing is now cloud storage, from which its consumers have reaped numerous benefits. Users can easily use public-key encoding and keyword searching to search secured data and retrieve desired data from cloud storage. However, there were some problems with speed, accuracy, and security when searching for encrypted keywords. The cross-lingual multi-keyword Centroid Merkle search over encrypted data (CLCMSE) suggested in this study is based on the Open Multilingual Wordnet and is meant to solve this problem. The proposed CLCMSE technique allows data users to query in any language and select the linguistic kind of information provided. First, the Centroid technique is used to cluster data from the cloud and sort the clustered data in this process. Then the Merkle search technique is used to increase the search speed. Finally, the targeted source data is retrieved from the cloud by using the fuzzy information retrieval algorithm. This experimental result proved that our proposed CLCMSE has higher accuracy, security, and speed performance than existing methods. This study compares the multi-keyword rank search over encrypted cloud data, the multi-keyword rank searchable encryption, and the verifiable attribute multi-keyword search.
Keywords:	Cross-Lingual Multi-Keyword, Centroid, Merkle Search, Cloud Storage, Word Net, Encryption, Data Retrieval.
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	A NEW MODIFIED GRAYSCALE IMAGE ENCRYPTION TECHNIQUE USING ELLIPTIC CURVE CRYPTOSYSTEM
Author:	ZIAD E. DAWAHDEH, MOHAMMED AMIN ALMAIAH, TAYSEER ALKHDOUR, ABDALWALI LUTFI, THEYAZN H. H. ALDHYANI, QUSAY BSOUL
Abstract:	Image encryption is one of the interested and important topics that recently spread as a result of the growing usage of the internet and other forms of communication in order to protect images from stealing and attacks. This work proposes a novel improvement to Menezes-Vanstone Elliptic Curve Cryptography to improve grayscale image encryption and decryption. The new modification in this paper reduces the encryption and decryption needed running time and speed up calculations. In the new method, no need for inverse and multiplication operations, only addition and subtraction are used, and this speeds up computations and reduces running time than other methods. Moreover, the modification makes the algorithm more secure and difficult for the attackers to attack it. Entropy, Unified Average Changing Intensity (UACI), and Peak Signal to Noise Ratio (PSNR) will be utilized to evaluate the grayscale image encryption efficiency. A comparison of the encrypted image and the original image will be performed to assess the performance of the suggested encryption approach.
Keywords:	Entropy, Unified Average Changing Intensity, Peak Signal to Noise Ratio, Elliptic Curve Cryptography, and Menezes-Vanstone Elliptic Curve Cryptosystem.
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text

Title:	OPPORTUNISTIC ROUTING USING RELAY NODE SELECTION IN ACOUSTIC WIRELESS SENSOR NETWORKS FOR EFFICIENT AND RELIABLE NETWORK PERFORMANCE
Author:	PRATHIBA N, MALA C S
Abstract:	Underwater Sensor Networks (UWSNs) represent a critical component of modern oceanic exploration and monitoring systems. These networks enable real-time data collection and communication beneath the ocean's surface, facilitating scientific research, environmental monitoring, and underwater surveillance. However, the exceptional challenges caused by the underwater conditions, such as limited bandwidth, high latency, and sporadic connectivity, demand innovative routing strategies to ensure efficient and consistent data transmission. This work emphases on the exploration of Opportunistic Routing (OR) in underwater sensor networks, a promising approach that leverages transient communication opportunities to enhance network performance. Opportunistic routing adapts to the dynamic and unpredictable underwater channel conditions, making it well-suited for UWSNs. It enables nodes to exploit intermittent connectivity and choose the most favourable transmission paths, thereby improving data delivery efficiency. Based on this concept of opportunistic routing, we have introduced a novel opportunistic routing protocol to improve the overall performance of UWSNs. The main aim of OR is to select the best suitable forwarding node which is chosen based on energy, node connectivity, and link quality. The link quality is estimated based on power; node connectivity analysis is done based on distance parameter. By evaluating these parameters, the final scheduling is assigned to accomplish the transmission task. The efficacy of proposed solution is validated through simulations where we have measured the performance in terms of throughput, packed delivery, packet loss and energy consumption. The comparison analysis shows that the proposed approach brought significant improvement when evaluated against with state-of-art schemes.
Keywords:	Underwater Sensor Networks, Opportunistic Routing, Relay Node Selection, Reliable Communication
Source:	Journal of Theoretical and Applied Information Technology 15^th April 2024 -- Vol. 102. No. 7-- 2024
Full Text