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

Welcome to Journal of Theoretical and Applied Information Technology

Submit Paper / Call for Papers

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

Journal of Theoretical and Applied Information Technology
January 2024 | Vol. 102 No.2

Title:	FROM THEORY TO PRACTICE: UNDERSTANDING THE FACTORS AFFECTING THE DEVELOPMENT OF DIGITAL COMMUNITY EDUCATION IN CHINA
Author:	SITI SARAH MAIDIN, LIU FAN, SIMON LAU BOUNG YEW
Abstract:	Digital community education in China is still in the development stage, with large regional differences and a low penetration rate. There is a lack of empirical studies on the influencing factors of the development of digital community education. This study investigates the use of digital community education online platform by Jinan residents, exploring the influencing factors, namely the residents use intention, education platform, education resources, team construction and the national government. A model of the influencing factors is established, and hypotheses about the relationships among the factors are made and validated using a structural equation model. The results provide a scientific basis for the proposed development strategy of digital community education, and recommendations are proposed for residents, platform resources, and national government to explore new paths for the development of digital community education. This study contributes to the understanding of digital community education development and provides practical implications for its improvement.
Keywords:	Digital Community Education, Influencing Factor, Empirical Study,
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	IMPROVING DEMAND FORECASTING FOR CONSUMER HEALTH PRODUCTS USING CLUSTERED LSTM MODELS
Author:	DINDA DIRA SEPTAMA, AMALIA ZAHRA
Abstract:	The study aims to enhance forecasting accuracy in a company's diverse product environment, focusing on two key objectives. First, this study aims to distinguish between clustered and non-clustered products in terms of forecast precision. Second, it investigates the possibility of utilizing aggregated forecasting models to more precisely predict product demand. Through clustering similar items and employing Long Short-Term Memory (LSTM) models, a notable improvement in demand forecasting accuracy was observed. Utilizing two years of consumer health product data and employing K-Means clustering, LSTM models tailored for each cluster outperformed non-clustered methods. Among 59 products grouped into 4 clusters, 20 demonstrated high (0-25% MAPE) and moderated (25-50% MAPE) forecast accuracy, surpassing only 9 products achieving similar precision without clustering. Further investigation into forecasting consumer health product demand is recommended. Additionally, the study explores the potential of creating aggregated forecasting models for efficiently predicting demand across multiple items.
Keywords:	Demand forecasting, LSTM models, Time-Series Forecasting, Clustering, Supply Chain Management)
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	PROPOSED SMART UNIVERSITY MODEL: THE INTEGRATION OF IOT AND FUZZY LOGIC IN SMART CLASSROOM FOR OPTIMIZING THERMAL COMFORT
Author:	ZAHRA OUGHANNOU, IBTISSAME KANDROUCH1, NOUR EL HOUDA CHAOUI, HABIBA CHAOUI, SALMANE BOUREKKADI
Abstract:	In the era of rapid urbanization, the emergence of "smart universities" provides a promising solution for the sustainable development of education. This article explores the transformative potential of integrating the Internet of Things (IoT) and fuzzy logic control in the education sector, with a specific focus on optimizing thermal comfort in classrooms. We begin by analyzing the fundamental principles of smart universities, emphasizing the central role of Information and Communication Technologies (ICT) in enhancing educational experiences. Drawing from various studies, we delve into the application of IoT in diverse educational contexts, emphasizing the creation of interactive learning environments and the rise of smart education. We present our proposed smart classroom architecture, taking into account the contextual considerations. Our Smart University model focuses on a crucial challenge: real-time analysis, aiming to propose an efficient architecture while considering the nature of technologies used within the Smart University. Our model is based on Fog Computing and Edge Computing to meet storage requirements, reduce communication load between Fog Nodes and sensors, and enable real-time analysis. Furthermore, we undertake an in-depth exploration of MQTT architecture, highlighting its significance in the client-server model. The article emphasizes the achievable improvement in thermal comfort through the fusion of IoT and fuzzy logic control, presenting a revolutionary approach to the management and optimization of classrooms. Through this comprehensive study, we envision a future where technology seamlessly integrates into the university landscape, offering sustainable and intelligent solutions to address contemporary challenges.
Keywords:	Smart University, IoT, Cloud Computing, Fog Computing, Wireless Communication, Fuzzy Logic, MQTT.
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	DESIGNING OF A WIRELESS CHARGING SYSTEM FOR ELECTRIC VEHICLES
Author:	B. MOHAN, M.V. RAMESH, RAJAN. VR, D. RENE DEV
Abstract:	Electric vehicles (EVs) are now commonly believed to be essential to the transition to smart transportation in the future. The biggest benefit is their contribution to lowering carbon emissions. Compared to conventional vehicles, EVs have advantages including reduced noise, economical maintenance, and zero pollution. But for EVs, the traveling range and charging process are major issues. The time required for charging the battery is more, and charging more vehicles at a time is difficult. According to this perspective, wireless power transfer (WPT) is a practical method for calculating EV trip distance and reducing the amount of time needed to charge the battery. So, in this proposed system a wireless charger allows an EV to automatically charge without the need for wires. To share electrical energy between two circuits by electromagnetic induction, a wireless charging device requires an electromagnetic field. The receiver coil was installed in vehicles below and the transmitter coil was fastened to the ground. When the main AC supply was given to the transmitter coil, it transferred electrical energy to the receiver coil through mutual induction, and with the help of converters the electric energy was given to the battery. The configuration of WPT to EV was simulated in MATLAB/Simulink and their physical parameters with mathematical calculations were analyzed through this proposed System.
Keywords:	Wireless Power Transfer, Electric Vehicle, Electromagnetic Induction, Transmitter Coil, Receiver Coil, Series-Series Compensation
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	HYBRID MODULATION TECHNIQUE TO IMPROVE RECEIVER SENSITIVITY FOR FSO LINK PROPAGATION
Author:	S. A. KWANG THAI, A. K RAHMAN, K. F. TAMRIN1, ENDUT. R, S. K. SAHARI
Abstract:	Free space optical (FSO) communication is now become a main communication due to the ability of propagation channel to operate up to Terabit per sec (Tbs) and can support high number user. The FSO suffer when experience severe weather condition. Apart from that FSO also facing threshold problem especially related with Amplitude Shift Keying - Onn Off Keying (ASK-OOK) when dealing poor signal and the biggest effect is high noise presence at receiver which led the signal to deteriorate. In this research proposed new development of transmitter and receiver design in order to reduce the impact of atmospheric attenuation and increase receiver sensitivity. In this paper focus on the analysis performance related bit rate which will compare with conventional amplitude shift keying (ASK) approach. Simulation result will be used to measure the performance and comparison between conventional and new proposed modulation double transmission balance receiver (DTBR) will also be presented. It was anticipated that proposed technique which offer a simple and inexpensive procedure capable of increasing received power, receiver sensitivity, and decreasing bit error rate would be used. The measurement of result will involve the effect of geometrical loss, data bit rate and distance propagation. Four level of synchronous transport module (STM) which is STM1(155Mbps), STM4(622Mbps), STM16(2.5Gbps) and STM64(10Gbps) will be compare the performance of bit rate. Meanwhile two different distances will test to measure the ability system extend the range transmission. From the result, the DTBR can increase 25% improvement as compare to conventional ASK.
Keywords:	Free Space Optical, Conventional ASK, Geometrical Loss, Bit Error Rate
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	AN INSIGHT INTO HYBRID AGILE SOFTWARE DEVELOPMENT APPROACH AND SOFTWARE SECURITY AMONG SOFTWARE ENGINEERS: A CRITICAL EVALUATION
Author:	SITI SARAH MAIDIN, NORZARIYAH YAHYA
Abstract:	The aim of this paper is to provide an overview of the hybrid agile development approach specifically in software development. Different platforms were used to investigate the factors that influence developers in choosing the preferred model for software development. In addition, this paper also examines the security elements in a software development project. It identified 3 factors that motivate software developers to focus on software security in a project. These are the company's policies and culture, i.e., the overall company culture regarding security, the application domain, i.e., the developers' perception of the benefits of security for their applications, and finally, the use and complexity of security tools. This article is organized as follows. The first section is the introduction, which explains the software development methodology. The subsequent Materials and Methods section provides an overview of searching journals in various databases, including Google Scholar, IEEE, ACM, and Science Direct. This paper concludes with recommendations on other areas that can be explored in the area of software security in the context of the hybrid agile approach.
Keywords:	Agile, Hybrid Agile, Software Development, Software Security, Model
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	IMPLEMENTING ADVANCED POWER QUALITY AND EFFICIENCY SOLUTIONS IN INDUSTRY 4.0: THE ROLE OF DYNAMIC LOAD BALANCING AND POWER FACTOR CORRECTION
Author:	YOUSSEF ZERGUIT, YOUNES HAMMOUDI, MOSTAFA DERRHI
Abstract:	The advancement of Industry 4.0 has ushered in a new era of smart industrial operations, driven by the integration of cutting-edge technologies. In this context, we present a pioneering solution that addresses the pivotal aspects of real-time dynamic load balancing and precision power factor correction in three-phase power systems. By harnessing the capabilities of Industry 4.0, our approach optimizes energy consumption while enhancing power quality. Our proposed system synergizes dynamic load balancing with efficient power factor correction in real-time, ensuring optimal distribution of loads and accurate compensation of power factors. This dynamic adaptation minimizes energy wastage and augments power utilization efficiency. The real-time nature of our solution empowers immediate adjustments, facilitating seamless response to varying load conditions. The key advantages of our approach encompass not only energy savings and enhanced power factor but also the integration of Industry 4.0 principles for predictive maintenance and data-driven decision-making. Consequently, industrial processes attain heightened responsiveness, cost-effectiveness, and environmental sustainability. In summary, this paper introduces an innovative contribution aligned with the Industry 4.0 paradigm. By providing real-time solutions for dynamic load balancing and power factor correction, our work enhances both energy efficiency and power quality, ultimately driving smarter and more resource-efficient industrial practices.
Keywords:	Automatic Balancing System, Power Factor Correction, Energy Efficiency, Power Quality, Three-Phase Loads, Three-Phase Balancing, Industry 4.0.
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	TOTAL QUALITY MANAGEMENT AND ITS ROLE IN DEVELOPING BANKING INSTITUTIONS THROUGH COMPLIANCE AS A MEDIATING FACTOR
Author:	SAMI ABDULLAH KADHIM, IBRAHIM RASOOL HANI
Abstract:	This study compared a random experiment to the data of a questionnaire Likert 7, for employees in Baghdad university, and their net number. The number of questionnaires obtained by the researcher was only (135) and was (126) answers correct, but was (6) missing and was (3) outliers. TQM scores were found to positively influence Banking Institution (BI) behaviors to determine the relationship between Total Quality Management (TQM) and Banking Institution (BI), r = 0.570, (r = 0.570, p < 0.05). The relationship between Total Quality Management (TQM) and Compliance (CO), r = 0.746, (r = 0.746, p < 0.05) and relationship between Compliance (CO) and Banking Institution (BI), can be used to improve employee behavior. r = 0.305, (r = 0.305 p < 0.05). This study describes how (TQM) on (BI).
Keywords:	Total Quality Management; Performance; Administrative Communications; Compliance; Goal Setting
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	RELIABILITY FOCUSED DESIGN PMIPv6 PROTOCOL WITH SECURE HANDOVER IN 5GC NETWORKS
Author:	MADHAVA RAO MAGANTI, DR K RAJASHEKAR RAO, DR. BALAJI VICHARAPU
Abstract:	Fifth generation (5G) networks deliver Massive Machine Type Communication (M2TC), better mobile broadband, and really dependable and minimal latency communications. To maximize these use cases, one must understand communication, 5G network segments, and architecture. These innovative network ideas require UE, RAN, and 5GC. Release 16 of the Third Generation Partnership Project included the Non-Access Level (NAL) and FG Application Protocol (FGAP) to improve RAN-5GC connectivity. A suggested outline supports reducing the conventional differences between EPC network components and improving flexibility inside the 5GC by ritualizing mobile network operations utilizing Reliability Focused Design PMIPv6 (REFDPMIPV6) in a cloud environment. The envisioned protocol defines protocol stacks and features pertinent to 5G networks, including resource allocation, data session formation, and authentication and identity processes. The protocol also discusses message flow related to Future Generation Node B (gNodeB) and UE registration. The suggested protocol exhibits resilience against a variety of assaults, successfully aligning with stated objectives, and has been rigorously modelled and validated using formal verification tools like BAN Logic and Scyther, as confirmed by simulation results.
Keywords:	5GC, FG-RAN, PMIPv6, REFD, NS3.
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	BLOCKCHAIN-ENHANCED CYBERSECURITY AND PRIVACY IN CLOUD COMPUTING: A SYSTEMATIC LITERATURE REVIEW
Author:	ISHRAG HAMID, MOUNIR FRIKHA
Abstract:	This paper presents a comprehensive exploration of blockchain-based privacy and cybersecurity solutions for cloud computing. Recognizing blockchain's origins in the realm of digital currencies, the authors meticulously explore its evolution into a robust, decentralized, and cryptographic tool, capable of addressing the intricate challenges faced in cloud computing environments. The study is comprehensive, covering not only the theoretical aspects of blockchain technology but also its practical applications, which extend beyond the traditional boundaries of cloud computing to include areas like digital currencies, smart contracts, and supply chain management. The authors critically assess the opportunities and challenges that blockchain integration presents for cloud computing. They shed light on how blockchain's inherent properties – such as immutability, transparency, and cryptographic security can effectively mitigate common security threats and privacy concerns in cloud environments. The review also acknowledges the complexities involved in implementing blockchain technology, such as issues related to scalability, energy consumption, and the need for regulatory frameworks that can adapt to the decentralized nature of blockchain. In essence, this paper offers a holistic view of blockchain's role in cloud computing, striking a balance between its potential to revolutionize data security and the pragmatic challenges that need addressing for its widespread adoption. The authors' exploration opens new avenues for future research and development in the field, highlighting blockchain's growing importance in the evolving digital landscape.
Keywords:	Blockchain, Cloud Computing, Cybersecurity, Data Privacy, Decentralization, Data Integrity, Cryptographic Algorithms, Scalability
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	BLOCKCHAIN-ENHANCED CYBERSECURITY AND PRIVACY IN CLOUD COMPUTING: A SYSTEMATIC LITERATURE REVIEW
Author:	ISHRAG HAMID , MOUNIR FRIKHA
Abstract:	This paper presents a comprehensive exploration of blockchain-based privacy and cybersecurity solutions for cloud computing. Recognizing blockchain's origins in the realm of digital currencies, the authors meticulously explore its evolution into a robust, decentralized, and cryptographic tool, capable of addressing the intricate challenges faced in cloud computing environments. The study is comprehensive, covering not only the theoretical aspects of blockchain technology but also its practical applications, which extend beyond the traditional boundaries of cloud computing to include areas like digital currencies, smart contracts, and supply chain management. The authors critically assess the opportunities and challenges that blockchain integration presents for cloud computing. They shed light on how blockchain's inherent properties – such as immutability, transparency, and cryptographic security can effectively mitigate common security threats and privacy concerns in cloud environments. The review also acknowledges the complexities involved in implementing blockchain technology, such as issues related to scalability, energy consumption, and the need for regulatory frameworks that can adapt to the decentralized nature of blockchain. In essence, this paper offers a holistic view of blockchain's role in cloud computing, striking a balance between its potential to revolutionize data security and the pragmatic challenges that need addressing for its widespread adoption. The authors' exploration opens new avenues for future research and development in the field, highlighting blockchain's growing importance in the evolving digital landscape.
Keywords:	Blockchain, Cloud Computing, Cybersecurity, Data Privacy, Decentralization, Data Integrity, Cryptographic Algorithms, Scalability
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	EVALUATION OF THE FETAL HEAD CIRCUMFERENCE FROM ULTRASOUND IMAGES DLA U-NET
Author:	MOHANA PRIYA.G, DR.S.K.B.SANGEETHA
Abstract:	This study introduces the DLA U-Net deep technique architecture in order to segment images method, which is specifically created for autonomously segmenting fetal ultrasonography images and head circumference (HC) biological dimension. To improve segmentation performance and accuracy, this method applies the consideration technique and deep supervision approach to U-Net representations. The evaluation of the proposed U-Net deep learning architecture (DLA) was conducted with the HC18 evaluation set, consisting of 355 cases. Four indicators of performance, namely the HC Mean Absolute Error (MAE), Hausdorff distance (HD), Dice similarity coefficient (DSC), and HC difference (DF) were used to assess the method's performance. U-Net deep learning architecture achieved a DSC of 97.94%, DF of 0.09 2.49 mm, MAE of 1.78 1.69 mm, and HD of 1.30 0.79 mm, according to experimental results. Gestational age (GA) was resolute by means of crown-rump length (CRL) dimension as the reference. GA estimations showed a mean difference of 0.5 ± 4.2, 0.3 ± 4.8 and 2.4 ± 12.3 days. The proposed U-Net architecture demonstrated superior or comparable segmentation performance when contrasted with contemporary techniques published in the literature. Hence, the introduced DLA U-Net can be considered as a viable approach for the delineation of fetal ultrasonography images and head circumference (HC) biological measurement.
Keywords:	Fetal ultrasound, DSC, Image segmentation, DLA, Head circumference, DF, Deep learning, U-Net
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	AN INTRUSION DETECTION APPROACH IN WIRELESS SENSOR NETWORK SECURITY THROUGH CNN-BI-LSTM MODEL
Author:	JIMSHA K MATHEW, KAVITHA NAIR R, B. KALPANA, MUTHULAKSHMI ARUMUGASAMY, S. SHARANYAA
Abstract:	Wireless Sensor Networks (WSNs) face growing safety threats, necessitating robust intrusion detection systems to safeguard network integrity. This research introduces a comprehensive framework for intrusion detection in WSNs, addressing the vulnerabilities inherent in these networks. The proposed approach integrates key techniques, namely data preprocessing through normalization, feature extraction utilizing Particle Swarm Optimization (PSO), and classification employing Convolutional Neural Networks and Bidirectional Long Short-Term Memory (CNN-Bi-LSTM).The framework commences with meticulous data preprocessing, wherein raw sensor data from the WSN undergoes normalization. This crucial step standardizes feature scales, ensuring data consistency and refining interpretability. Subsequently, the PSO algorithm is applied for feature selection, optimizing the identification of relevant features. By minimizing redundancy and maximizing the discriminative power of the feature set, PSO significantly enhances intrusion detection capabilities.The selected features serve as input for the CNN-Bi-LSTM model, a powerful combination leveraging CNN's spatial feature extraction and Bi-LSTM's temporal modelling. CNN captures high-level spatial representations from the input features, while Bi-LSTM effectively captures temporal dependencies in sequential sensor readings. This synergy equips the framework to discern complex intrusion patterns with heightened accuracy.Performance evaluation, conducted using labelled datasets, demonstrates the superior efficacy of the integrated framework compared to other intrusion detection methods. The experimental results underscore the framework's remarkable ability to achieve enhanced intrusion detection performance in WSNs, solidifying its significance in advancing the security paradigm for these critical networks.
Keywords:	Intrusion Detection, Wireless Sensor Networks, Convolutional Neural Networks, Bidirectional Long Short-Term Memory, Particle Swarm Optimization
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	DYNAMIC SLICING OPTIMIZATION IN 5G NETWORKS USING A RECURSIVE LSTM MECHANISM WITH GREY WOLF OPTIMIZATION
Author:	MOHAMMED ASHFAQ HUSSAIN, AHMED UNNISA BEGUM, ZIAUDDIN SYED, DR. MOHAMMED SALEH AL ANSARI
Abstract:	5G networks provide unmatched speed, connectivity, and the ability to support a wide range of diversified services, ushering in a transformational era of telecommunications. By applying innovative methods in data analysis, time-series forecasting, and optimization, this work provides a thorough strategy to addressing these difficulties. Detailed configuration and performance management data collecting from a 5G experimental prototype forms the basis of our methodology. Slicing ratios, priority, QCI (Quality of Service Class Identifier), and power measurements are among the critical parameters included in this collection. This work uses min-max normalization to guarantee consistency and standardized scaling in order to get this data ready for in-depth examination. For time-series forecasting, this novel method presents the Recursive LSTM (Long Short-Term Memory) model. LSTM networks, which are well-known for their ability to capture long-term dependencies, are essential for identifying temporal patterns in the information. In order to carefully adjust parameters and improve dynamic slicing configurations, this work employs the Grey Wolf Optimization (GWO) method. The GWO algorithm makes sure that network resource allocation constantly adapts to fulfill various objectives, taking inspiration from the hierarchical grey wolf pack's structured decision-making process. The combination of these advanced techniques results in a solution that greatly improves the accuracy and flexibility of time-series forecasting and resource distribution in 5G networks. Through the harmonic integration of data-driven insights, LSTM predictions, and the effective optimization capabilities of GWO, our methodology enables 5G networks to allocate resources with agility and flexibly, ultimately providing real-time, high-quality services. The method outperforms other approaches like CNN, CNN-LSTM, and RNN-LSTM, which were all implemented with MATLAB, by a substantial margin of 5.55%, with an accuracy of 99.12%.
Keywords:	5G Network, Quality of Service Class Identifier, Long Short-Term Memory, Grey Wolf Optimization, Min-Max Normalization.
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	TRANSMIGRATION 4.0: IMPLEMENTATION OF INTEGRATED FUNCTIONS AND TECHNOLOGY
Author:	SURATMAN, GERARDA ORBITA IDA CAHYANDARI, SRI RAHAYU BUDIANI, SERI ARYATI
Abstract:	Transmigration has been a complex program conducted by the Indonesian government since the colonial period. The complexity is related to economic, political, social, and cultural aspects. Government must tackle the issues and problems in transmigration. The changing goals of transmigration through periods have influenced the programs at all costs. This study proposes the implementation of transmigration 4.0 in Indonesia. The new concept of transmigration in the Industry 4.0 era emerges as the integration of functions and connections on the macro scale and micro scale through developing area units. The program adheres to national and regional development planning. Two types of data were collected for this study: primary and secondary. Interviews, focus group discussions (FGD), and other technical field surveys were used for primary data collecting. The outcome demonstrates that agricultural primary products, which are cashews, coconut, and corn, can improve economic productivity. The cultivation and production require the participation of migrants and local people. The process needs an integrated place called Trans Science Techno Park (TSTP). TSTP is an agricultural science techno park located in a transmigration region. Transmigration region in Muna Regency, Southeast Sulawesi Province, will have TSTP as pentahelix collaboration consisting of ICT media for digital marketing and branding, research and development, community, government, and industry. The convergence of technology and agriculture in TSTP is known as Transpolitan 4.0 and is replicable in other contexts. The core of Transpolitan is intertwined with agroproduction, agroindustry, agribusiness, agrotechnology, and agrotourism.
Keywords:	Transmigration, Transpolitan, Science Techno Park, Agriculture, Pentahelix
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	ADVANCEMENTS IN INTRUSION DETECTION SYSTEMS FOR INTERNET OF THINGS: A STATE-OF-THE-ART AND COMPREHENSIVE ANALYSIS OF MACHINE LEARNING ALGORITHMS
Author:	RACHID HDIDOU, MOHAMED EL ALAMI
Abstract:	The Internet of Things is one of the technologies that form the basis of the modern technological revolution. The use of this technology in various fields has become a necessity due to development and the speed with which the world is changing. This use depends on and is linked with the resolution of some issues that impact the technology of the Internet of Things, among these issues, is the issue of security. Internet of Things computer security is considered one of the important points in the sustainable and secure use of Internet of Things technology in most fields, especially sensitive fields such as the medical field, the military field, the banking field, and other fields. Intrusion detection systems are considered one of the appropriate techniques to secure networks and IoT applications due to their flexibility in application. However, with the great development of cybercrime, the standard solutions of intrusion detection systems have become insufficient to secure applications and networks of the Internet of Things. This confirms the need to propose and develop intrusion detection system solutions based on artificial intelligence and machine learning techniques. In this paper, we will present a state-of-the-art and analytical study on Internet of Things security using intrusion detection system solutions based on Machine Learning algorithms.
Keywords:	Intrusion Detection Systems, Internet of Things, Machine Learning Algorithms, IDS, IoT.
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	FACTORS SHAPE THE SUCCESS OF IMPLEMENTING AR MODELS FOR STEM LEARNING
Author:	HANI DEWI ARIESSANTI, HARJANTO PRABOWO, FORD LUMBAN GAOL, MEY LIANA
Abstract:	Research on AR in STEM education requires information about AR models and their impact on student learning. In higher education, many studies have been conducted on the use of AR in STEM education to determine its impact on student learning. However, there is a knowledge gap and little research on the application of AR-STEM education in middle schools. Furthermore, the impact of AR-STEM education on student learning has not been studied. To address the knowledge gap, this article reviewed 42 articles in the Methods and Meta-Analyses Reference Manual – Assessment (PRISMA) using a new systematic literature review (SLR). Research on student learning outcomes in STEM education typically includes six categories of variables that influence learning outcomes (e.g., cognitive knowledge, understanding of STEM material, technology usability, knowledge synthesis ability, visualization of virtual objects). However, this study will only produce three variables including use of technology, synthesis ability and cognitive ability. The contribution of this research helps professionals and teachers to provide student learning experiences through the AR platform environment.
Keywords:	Student Learning Outcomes; AR For STEM Learning; STEM Learning; PRISM; Systematic Review
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	PROSTATE CANCER DETECTION USING GREEDY SEARCH FOR LESIONS IN MAGNETIC RESONANCE IMAGES (MRI): A NOVEL TECHNIQUE
Author:	GUNDLOORI MUBARAK , RAJABHUSHANAM C
Abstract:	Prostate cancer is one of the leading causes of cancer death in males. The death rate from prostate cancer can be drastically lowered with well-thought-out plans for treatment and maintenance. Cancer diagnosis by magnetic resonance imaging is a great first step in treatment (MRI). New research shows that MRI can also be used to categorise prostate cancer. There may be hereditary cancer implications for men and their families because genetic abnormalities can greatly raise the risk for prostate cancer (PC), and may be associated with aggressive disease and poorer outcomes. MRI is a useful tool for diagnosing the pathological conditions associated with PC (MRI). Swarm intelligence is the basis for Particle Swarm Optimization (PSO), an algorithm used to solve optimization problems in search spaces and to analyse and forecast social behaviour under the assumption of the existence of goals. In order to facilitate PC detection in MRIs, the authors herein employ a greedy technique for hybridising PSO optimization. The Adaboost classifier is another option that has been proposed. In order to better diagnose prostate cancer, this study suggests using an internal learning technique.
Keywords:	Prostate cancer diagnosis, MRI imaging, object recognition, inner learning, and prostate segmentation
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	THE MODEL OF LOCAL WISDOM FOR SMART WELLNESS TOURISM WITH OPTIMIZATION MULTILAYER PERCEPTRON
Author:	SUNU JATMIKA , SYAAD PATMANTHARA , AJI PRASETYA WIBAWA , FACHRUL KURNIAWAN
Abstract:	This study focuses on the influence of variations in the number of hidden layers in the Artificial Neural Network (ANN) method on model performance and interpretability of results. The method applied involves integrating local wisdom to optimize the Artificial Neural Network (ANN) model. This approach combines locally relevant aspects with a conceptual framework to improve ANN performance. Evaluation of the results involves the performance metrics, MSE, MAE, RMSE, and F2 Score to find the best-hidden layer pattern in the Artificial Neural Network (ANN) model. The test results are based on a dataset with five indicators totaling 30 input layers and tested on the Multi Layer Perceptron (MLP) model. The results of testing a dataset with 30 input layers divided into 5 indicators produced performance metrics MSE 0.01346, MAE 0.09740, and RMSE 0.12094. The concept with a 16-hidden layer model pattern has high complexity and produces better predictions with fewer errors. Additionally, hidden layer 11 performs well, displaying a solid capacity to describe the variance in target data with an R2_Score of 0.17374. This produces two groups of ANN test results: the first group with improved accuracy (MSE, MAE, RMSE), and the second group highlights the optimal performance of hidden layers 16 and 11 (R2 Score). Local wisdom contributes to smart wellness.
Keywords:	Artificial Neural Network, Local Wisdom Integration, Performance Metrics, Hidden Layer Patterns, Multi Layer Perceptron
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	AUTOMATIC GENERATION OF MULTIPLE-CHOICE QUESTIONS USING TEMPLATE-BASED SEMANTIC WEB IN INDONESIAN LANGUAGE
Author:	SHARFINA FAZA , ROMI FADILLAH RAHMAT , DANI GUNAWAN , RINA ANUGRAHWATY , SILMI , GABRIELA DWI LADY , FARHAD NADI
Abstract:	Questions serve as tools to assess a student's knowledge and understanding. However, creating questions on a large scale can be time-consuming. To address this issue, an automated question generation system offers a solution. An automatic question generation system, also known as a question generation system, is an application that automatically generates questions from a given text or document using a specific method. This study applies a template-based method using semantic web to generate questions categorized as non-factoid and factoid. In order to implement semantic web, the system utilizes ontology modeling, which includes main components such as classes, properties, and instances. The ontology modeling process is carried out using a software called Protégé. The dataset used in this study consists of biology and history subject matter for high school students, obtained from subject matter providers on the web. The system checks the dataset for keywords, categorizes the questions, and places them into the appropriate question template. The automated question generation system successfully generated a total of 2468 questions. During the testing phase, 2132 questions were deemed acceptable according to their respective categories, while 336 questions were not accepted. By calculating the percentage of accepted questions over the total questions, the overall accuracy rate was found to be 86.38%.
Keywords:	Questions, Automated Question Generation System, Template-Based Method, Semantic Web.
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	A NOVEL HEURISTIC FOR GRAPH-BASED TOPIC MODELING USING SPECTRAL CLUSTERING
Author:	P.K.PATTANAYAK, R.M.TRIPATHY, S.PADHY
Abstract:	Topic modeling is one of the popular techniques for identifying the latent topic from a large corpus of text data. Various topic modeling techniques have been studied for managing short and long texts that consider different kinds of interactions and constraints within a dataset. Most researchers use Latent Dirichlet Analysis (LDA) and an extension of the LDA algorithm for topic modeling. While these algorithms are flexible and adaptive, they are occasionally a poor choice for modeling increasingly complex data relationships. Topic modeling has used various encoding strategies, many of which do not adequately represent the semantic relationships between the words. This study proposes a novel heuristic for the graph-based topic modeling technique and applies it to a benchmark dataset, which outperforms the current LDA model for short text. The proposed heuristic is based on graph-splitting methods. We used it on the TripAdvisor hotel review dataset, a sizable collection of huge text corpora. Our suggested strategy has been demonstrated to outperform several current methods for concept extraction and effective topic. The detailed result was compared based on the coherence score. We also employed word cloud and compared the outcome to user reviews, demonstrating that our performance is superior to many of the already used methods.
Keywords:	Topic Modeling, Spectral Clustering, Word2Vec, Graph-Splitting, Linear Dirchlet Allocation
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	A NEW CLASS OF TWO-DIMENSIONAL (2-D) MODIFIED-FCC (M-FCC) OCDMA FOR CARDINALITY ENRICHMENT TOWARDS 6G NETWORK ACCESSIBILITY
Author:	H. DAYANG, C.B.M RASHIDI, S.A ALJUNID
Abstract:	In this paper, we introduce a new two-dimensional (2-D) wavelength/time optical code division multiple access (OCDMA) known as 2D Modified-Flexible Cross Correlation (M-FCC) code. The 2-D M-FCC code is developed from the one-dimensional flexible cross correlation code families with flexibility in-phase cross-correlation at any given number of users and weights. The main goal is to reduce noise as well as multiple-access interference (MAI) at the same time to accommodate higher cardinality with minimum noise. The increment of cardinality is to be engaged towards 6G network accessibility. From the numerical results, it indicated good performance whereas the 2-D M-FCC code at bit rate 622 Mbps with at BER of 3.49 ×10-28, can accommodate 200 number of cardinalities enrichment. Moreover, the 2-D M-FCC code shows good SNR curves as gradually decreases as the number of users increases at different bit rate 155 Mbps, 622 Mbps, 1.1 Gbps and 2.5 Gbps, respectively.
Keywords:	2D CODE, OCDMA, MAI, SNR, BER
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	FAULT DETECTION IN DIESEL ENGINES USING ARTIFICIAL NEURAL NETWORKS AND CONVOLUTIONAL NEURAL NETWORKS
Author:	RIZQI FITRI NARYANTO, MERA KARTIKA DELIMAYANTI, AGUSTIEN NARYANINGSIH, RIZKY ADI, BIMA AJI SETIAWAN
Abstract:	This research aims to create a deep learning model utilizing Artificial Neural Networks (ANN) and Convolutional Neural Networks (CNN) to detect and classify damage or faults in diesel engines effectively. The main contribution of this study entails the creation of a classification model aimed at problem identification in diesel engines. The study used the DEFault dataset, which has 3,500 rows of data classified into four distinct labels. The DEFault dataset comprises four discrete noise levels, namely 0dB, 15dB, 30dB, and 60 dB. The findings indicate that both models have demonstrated satisfactory outcomes regarding model assessment. The model performance is most ideal when trained on the DEFault dataset with 60dB white noise, whereas the dataset with 0dB white noise leads to the least favourable model performance. The results suggest that artificial neural networks (ANN) perform better than convolutional neural networks (CNN) in datasets with higher levels of white noise. Conversely, CNN demonstrates superior performance in datasets with lower levels of white noise. An additional investigation could prove advantageous in implementing the deep learning model on a device that can identify diesel engine faults in real-time.
Keywords:	Diesel Engine, Fault Detection, Deep Learning, Artificial Neural Network (ANN), Convolutional Neural Network (CNN)
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	COLLABORATIVE COMMERCE MODEL IN MSMES AND CREATIVE INDUSTRIES IN PAREPARE: CO-DESIGNING SUSTAINABLE PRODUCT INNOVATION
Author:	KHAERA TUNNISA , WAHYUNI EKASASMITA , ANDI MUHAMMAD BAHRUL ULUM
Abstract:	In response to the impact of the environmental issue of plastic waste, which is changing customer shopping patterns and product consumption patterns, MSMEs and Parepare Creative Industries are collaborating using Collaborative Commerce (C-Commerce). MSMEs have carried out various collaboration efforts but have yet to evaluate products and customer feedback. Therefore, this research was conducted to build a collaborative commerce conceptual model for MSMEs and Creative Industries as a collaboration standard. Data collection was carried out through observation, interviews, and literature study. The sample selection used purposive purposes, specifically local food products and typical Parepare handicrafts. Soft System Methodology (SSM) is based on a conceptual model in C-Commerce. SSM is a suitable method for building a new ecosystem. The conceptual model is compared with the rich picture for a complete and holistic understanding. The result of this conceptual model is to create a C-Commerce system for MSMEs and the Parepare Creative Industry to design sustainable products and a platform-based ecosystem. This concept is appropriate for Parepare MSMEs and Creative Industries to maintain cooperative relationships and increase product flexibility.
Keywords:	Collaborative Commerce, MSMEs, SSM, Information Technology, Sustainable Product
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	THE USE OF NUMERICAL ANALYSIS FOR SCALE ANALYSIS AND NUMERICAL INTEGRATION BASED ON SCALE
Author:	MUNA JAAFAR RAHEEM
Abstract:	For a broad numerical method to calculate the measure function, we give a convergence analysis. We suggest a particular approach for estimating the measure functional and examine the convergence ratio by combining Lagrange extrapolation. Additionally, we examine the numerical measure integration error bound and demonstrate how it can reduce singularity for singular integrals. Theoretical findings are supported by numerical examples. All of this will be studied in detail through mathematical equations and laws in this research, so that we will gradually clarify and introduce the approved method.
Keywords:	Numerical Integrating Method, Numerical Measurement Integrated (NMI), Lagrangian Interpolation, Lebesgue Integrated Error, Scal Analysis
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	WHEAT CROP GROWTH PARAMETERS RETRIEVAL USING FULL POLARIMETRIC GROUND-BASED SCATTEROMETER DATA
Author:	GEETA T. DESAI , ABHAY N. GAIKWAD
Abstract:	This article presents the utility of radar vegetation index (RVI), Pauli decomposition and full polarimetric data acquired, using a ground-based scatterometer to retrieve canopy height, leaf length and stem width of wheat crop. Polarimetric backscatter data at L and S bands at different incidence angles from 100 to 700 in steps of 50 was acquired over phenological stages of the wheat crop. In situ ground data from the field, including canopy height, leaf length and stem width, was collected to validate results. The RVI and Pauli decomposition components were computed, and their contribution to wheat parameter estimation was further explored using generalized regression neural network. Pauli decomposition has been utilized to investigate the performance of full polarimetric multitemporal scatterometer data for monitoring growth stages of wheat crop. Comparing observed and predicted canopy height, stem width, and leaf length retrieved from polarimetric data yielded root mean square error (RMSE) values varying from 1.327 to 0.014. The study revealed that the best retrieval results were obtained after combining backscattering values with RVI. This combination resulted in RMSE values of 0.560, 0.179 and 0.014 for canopy height, leaf length and stem width, respectively.
Keywords:	Ground Based Scatterometer (GBS), Canopy Height (CH), Radar Vegetation Index (RVI) and Generalized Regression Neural Network (GRNN).
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	HARDWARE/SOFTWARE CO-DESIGN IN MULTIPROCESSORS EMBEDDED SYSTEMS AND IOT
Author:	KAMEL SMIRI, RADIA BEN DIMERAD, FAYCEL EL AYEB
Abstract:	This article presents a comprehensive literature review of current research focusing on the design of Multiprocessor Embedded Systems (MPSoC) and their integration with the Internet of Things (IoT). Additionally, it investigates the prevalence of software/hardware co-design methodologies aimed at enhancing the overall performance of these systems. IoT and MPSoC stand as significant research paradigms, and the incorporation of hardware/software co-design emerges as a pivotal strategy to expedite the processing and analysis of data flows. The article not only defines hardware/software co-design but also underscores its critical role in optimizing the functionality of IoT and MPSoC. Furthermore, it provides clear definitions of IoT and MPSoC along with a comprehensive exploration of various proposed architectures based on these foundational concepts. In conclusion, the article puts forth future research directions, delving into the potential implications of co-design in the context of IoT and MPSoC. It also considers the integration of emerging technologies such as cloud and fog computing, machine learning, and advanced mobile systems like 5G/6G. This forward-looking approach aims to guide future developments in the dynamic intersection of co-design, IoT, and MPSoC.
Keywords:	Internet of Things; Embedding; Hardware/Software Co-Design; Engineering Design Problem.
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	IMBALANCED CLASS LEARNING IN VISION BASED CLASSIFICATION OF VECTOR MOSQUITO SPECIES
Author:	RESHMA PISE , KAILAS PATIL
Abstract:	Vector-borne diseases, primarily transmitted by mosquitoes, remain a significant global public health concern. Accurate and timely identification of mosquito species is crucial for comprehending disease transmission patterns and implementing effective vector control measures. In recent years, vision-based deep learning techniques have shown promising results in the classification of mosquito species. However, the natural class imbalance present in real-world mosquito species datasets can negatively impact the predictive performance of CNN based classifiers. This paper presents three popular class imbalanced learning strategies: Oversampling, Under-sampling, and Synthetic Minority Oversampling Technique (SMOTE) to address the skewed class distribution. We investigate the effectiveness of these solutions on the imbalanced images dataset of vector mosquito species with the aim of enhancing the performance of CNN-based classifier. The classification outcomes demonstrate that SMOTE based CNN outperforms other techniques in terms of evaluation metrics: sensitivity, specificity, F-score and accuracy. Class imbalance learning techniques are vital in vector control applications where the accurate classification of rare class i.e., harmful species is crucial for effective monitoring and control of disease vectors.
Keywords:	Class Imbalance, Deep Learning, Mosquito Species, Image Classification, Sampling Techniques.
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	ENHANCING ANDROID SECURITY: NETWORK-DRIVEN MACHINE LEARNING APPROACH FOR MALWARE DETECTION
Author:	HUSSEIN A. AL-OFEISHAT
Abstract:	The exponential growth of malevolent Android applications poses a significant threat to the security of computer networks. This study presents a machine learning approach utilizing network-based techniques to enhance the detection of malware, thereby enhancing the security of Android systems. The study encompasses various stages, including dataset preprocessing, feature engineering, model building, evaluation, and application. The dataset encompasses a diverse range of Android applications that have been classified as either malicious or benign. Feature selection methods are employed to identify pertinent characteristics at the network level, thereby enhancing the performance of the model. Precision, recall, F2 score, and average precision (AP) are among the metrics employed to evaluate the effectiveness of the model. The findings demonstrate that the optimized model exhibits superior performance compared to the baseline methods, effectively identifying Android malware with minimal occurrences of false positives and false negatives. The implementation of the model within a network security environment in the real world serves as evidence of its validity for potential integration into future systems. The present study offers a substantial contribution to the domain of network security by demonstrating the efficacy of integrating machine learning techniques with network-level characteristics to enhance the effectiveness of virus detection on the Android platform. The aforementioned findings underscore the necessity for ongoing research in this domain to remain abreast of emerging risks and enhance network safeguards against Android malware.
Keywords:	Android Security, Malware Detection, Machine Learning, Network-Driven Approach, Feature Selection, Network Security.
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	SMART HUSBANDRY FEEDING SYSTEM USING IOT AND MOBILE APP’S TO DETERMINE THE MOST FEEDING ESTIMATION AND PROFIT MAXIMIZATION
Author:	MUHAMMAD IMAN SANTOSO , ANGGORO SURYO PRAMUDYO , MUHAMMAD IRVAN MAULANA , MUHAMMAD FAKHRURIZA PRADHANA
Abstract:	Chicken farms are typically situated far away from residential areas to prevent conflicts, pollution, and noise. However, this distance results in increased time and costs for managing remote coops. Effective control over chicken feeding is crucial for efficient farm management. An automated approach is necessary, as manual feeding is more wasteful, lacks regularity, and involves higher labor costs. This study proposes a smart feeding system that utilizes IoT and Mobile Apps. The key components of the device include Nodemcu, load cell sensor, DHT11 sensor, relay, RTC DS3231, servo motor, and I2C 16x2 LCD. Simultaneously, the mobile application manages various primary features, such as temperature, humidity, light settings, feed information, chicken quantity and age settings, and an auto/manual feed schedule. The system underwent a series of tests, and the assessment confirmed that the system can monitor data, manage feed, and regulate temperature of the coop using a lamp in real-time. Additionally, the mobile application allows users to access the feed history based on the feeding date. The remote-control system for feeding, temperature, and humidity offers reduced operational costs and helps maintain a safe distance between poultry and humans. The research results present significant implications as they streamline automated feeding processes for poultry farmers and introduce a precise algorithm for calculating feed requirements. The proposed tool and algorithmic system not only demonstrate cost-effectiveness but also showcase efficiency gains in terms of time and labor. This product is particularly well-suited for small and medium-sized enterprise (SME) chicken farmers, providing valuable contributions to the advancement of automated feeding systems within the poultry industry. It fosters enhanced management practices and economic optimization.
Keywords:	Smart Feeding System, IoT, Husbandry, Mobile App’s, Profit Maximimization
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text

Title:	UNRAVELING LUNG CANCER THROUGH GENOMIC INSIGHTS AND ENSEMBLE DEEP LEARNING
Author:	K. MARY SUDHA RANI, Dr.V. KAMAKSHI PRASAD
Abstract:	The exponential growth in genomic data availability has spurred innovative cancer prediction strategies In this study, we applied "Gene Set Enrichment Analysis (GSEA)" alongside potent deep learning techniques to forecast lung cancer. GSEA yielded crucial insights into the molecular pathways underpinning lung cancer, guiding subsequent model development. Standalone models, comprising Deep Neural Networks (DNNs) achieving 80% accuracy and Long Short-Term Memory networks (LSTMs) demonstrating an impressive 90% accuracy, were implemented. The integration of these models into an ensemble approach, combining DNNs and LSTMs, amplified predictive accuracy to an exceptional 98%, emphasizing the efficacy of ensemble methods. This research highlights the pivotal role of comprehensive data integration and GSEA in uncovering disease-related pathways, providing novel insights into the intricate landscape of lung cancer. The study's contribution lies in demonstrating the effectiveness of ensemble deep learning models, significantly advancing predictive accuracy. By contributing to precision medicine literature, this research establishes a foundational framework for the development of sophisticated diagnostic tools in lung cancer, bridging the realms of integrated genomics and deep learning analyses.
Keywords:	Gene Set Enrichment Analysis (Gsea), Dnn, Lstm, Ensemble Deep Learning, Lung Cancer Prediction, Precision Medicine
Source:	Journal of Theoretical and Applied Information Technology 31^st January 2024 -- Vol. 101. No. 2-- 2024
Full Text