		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

October 2023 | Vol. 101 No.20

Title:	SPEED UP THE DEEP BIDIRECTIONAL TRANSFORMERS WITH FEATURE SELECTION FOR SUMMARIZING MEDICAL PAPERS
Author:	ALSHIMAA.M.IBRAHIM, MOSTAFA MAHMOUD AREF, MARCO ALFONSE
Abstract:	Medical papers are being widely published currently, especially after the Coronavirus disease (COVID-19) pandemic. The time required to manually summarize medical papers can be decreased by applying text summarization approaches. It is now common practice to overcome medical text summarization challenges using pre-trained models such as the Bidirectional Encoder Representations from Transformers (BERT)-base model. This paper presents a new system for summarizing medical papers based on deep learning techniques. In this system, we combine the -Statistic (CHI-square) feature selection technique with a token classification such as Part-of-Speech (POS) tagging and use the feature selection output as input to the pre-training BERT-base model, then apply clustering algorithms for the sentence selection process. Our main contribution is that our model obtained high speed and accuracy compared to previous summarization methods. We performed our comprehensive experiment on the public corpus that was randomly selected from BioMed Central (BMC). In comparison to other models that need a lot of training time, our model's output has high performance and is less complex. We used the Recall-Oriented Understudy for Gisting Evaluation (ROUGE) metric for an evaluation process. The model results are ROUGE-1 = 0.7611, ROUGE-2 = 0.3205, and ROUGE-L=0.4544.
Keywords:	Text Summarization, Machine Learning, Deep Learning, Natural Language Processing, Feature Selection.
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	MALICIOUS NODE FEATURE SELECTION USING SWARM INTELLIGENCE CLASSIFIER IN WIRELESS SENSOR NETWORK
Author:	VEMULA KESAVA KUMAR, P. SURESH VARMA
Abstract:	Rapid development and growth of technology Various real-time applications are used in wireless environments. Internet-enabled services are the most important in the world for the delivery of efficient services. There are a variety of quality factor issues and security attacks in wireless sensor network environments. Network efficiency is measured to find the attacks and malicious node functions. In this white paper, we propose efficient detection of malicious nodes and their function using deep learning. A novel swarm intelligence method is used to measure the features and apply a classification technique to measure the gain. In this document we used the dataset Knowledge Discovery Dataset UCI Repository to calculate the performance. The identifier is calculated using Tensorflow. The various functions evaluate the performance of accuracy, detection time, turnaround time, energy consumption and packet delivery ratio. The system accuracy proposed by us is to be calculated and the characteristics compared with existing methods.
Keywords:	Energy Efficiency, Malicious Attack, Swarm Intelligence, Tensorflow, Wireless Sensor Network.
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	EXPLORING ELECTRONIC SYSTEM MODERNIZATION AND DIGITAL LITERACY ABILITY: COMPARATIVE LITERATURE REVIEW
Author:	IRIYADI, MEIRYANI, ROBIATUL HIDAYAH, HERI SUKENDAR WONG, KRISWANTO, ARMANTO WITJAKSONO, SASYA SABRINA, TRIANA MEINARSIH, SEPTI WIFASARI
Abstract:	This research aims to discuss modern electronic systems associated with digital literacy with a comparison of literature reviews related to this research. This research uses qualitative methods with secondary data collection techniques from scopus-indexed journals and local journals that have been officially published. In this research, we explore the development of digital literacy in different countries and the Indonesian government's efforts to combat fake news in the digital age. The study reveals various approaches to enhancing digital literacy worldwide, emphasizing the importance of adapting to local contexts. Notably, advanced technologies such as virtual reality and artificial intelligence play significant roles in improving digital literacy. The research also underscores the vulnerability of individuals to fake news and its consequences, which necessitates government action. Recommendations from this study include the promotion of literacy, both digital and conventional, to empower individuals to discern accurate information from misinformation. Government initiatives should focus on organizing literacy campaigns and expanding online education opportunities to ensure a more informed and resilient society. This research sheds light on the critical role of literacy in an ever-evolving digital landscape, offering valuable insights for individuals, communities, and governments.
Keywords:	Digital Transformation, Media Literacy, Educational Innovation, Digital Literacy, Modern Electronic System.
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	APPLICATION OF SENTIMENT ANALYSIS IN SOCIAL NETWORKS: A CASE OF ANALYZING ONLINE HOTEL REVIEWS
Author:	SARA ALCABNANI, FAYÇAL ZINEDINE, SOUAD EL HOUSSAINI, JAMAL EL KAFI
Abstract:	In a world where competition is fierce, companies try to acquire a good reputation with their customers. Electronic reputation is a part of this reputation especially in the context of social networks, where everyone can freely express their opinion. Social networks are increasingly being used in the hotel industry. As hotels operate in a competitive and dynamic environment, it is essential that they make effective use of online customer review information to better understand their customers, improve their performance and compete with other hotels. The availability of vast amounts of user-generated data on social networks has led to a growing interest in using automated computational methods such as text mining and sentiment analysis to process large amounts of user-generated data and extract meaningful knowledge and insights. The objective of this work is to explore the opinions of internet users on a set of hotels in Morocco. Our study will contribute to the literature on social network analysis by uncovering rich new findings and providing actionable insights with implications for hotel managers in Morocco. Studies show remarkable performance gains for companies that seize the opportunities offered by analytics Therefore, the purpose of this paper is to show companies how they can investigate and improve their E-reputation by integrating emerging technologies to analyze unstructured textual data available on social networks using a case study. Our study is divided into several parts: data acquisition allowing to conduct a study based on machine learning, preprocessing, where we filtered this data (eliminate unnecessary words, use tokenization ...) to keep the information needed for better accuracy, the application of Machine Learning algorithms (SVM, Naive Bayes, Decision Tree, Random Forest, and logistic regression) for a supervised classification where the results are binary (positive/negative) and finally the development of an application that allows the realtime E-reputation monitoring. The solution supports two languages: standard Arabic and French
Keywords:	NLP, Machine learning, Sentiment analysis, Opinion detection, E-reputation
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	MULTI-AGENT PROXIMAL POLICY OPTIMIZATION FOR PORTFOLIO OPTIMIZATION
Author:	FIRDAOUS KHEMLICHI, HIBA CHOUGRAD, SAFAE ELHAJ BEN ALI, YOUNESS IDRISSI KHAMLICHI
Abstract:	Deep reinforcement learning is a subfield of machine learning that combines the ideas of deep learning and reinforcement learning to enable agents to learn and make decisions in complex environments. It has been applied to a wide range of tasks, including gaming, robotics, and finance, among others. In finance, reinforcement learning (RL) has emerged as a promising technique for solving strategic decision-making problems in complex financial environments using reward-based approaches for optimal control. In this paper, we propose a novel algorithm that leverages the power of Multi-Agent Reinforcement Learning (MARL) coupled with Proximal Policy Optimization (PPO) to tackle the complex problem of portfolio optimization. What sets this approach apart is its utilization of MARL, which involves multiple agents learning and interacting within the same environment. This is in contrast to the traditional single-agent approaches commonly used in portfolio optimization. In portfolio optimization, MARL enables agents to learn from the interactions with other agents and the environment, leading to more realistic and robust investment strategies. The performance of the algorithm was assessed on the S&P 500 market using various numbers of agents and assets, and its performance was compared to several benchmarks. The performance metrics used for evaluation consisted of annual profit, annual volatility, Sharpe ratio, and Sortino ratio. The findings demonstrated that the algorithm outperformed the benchmarks in terms of all the performance metrics considered, regardless of the number of agents and assets involved.
Keywords:	Deep Learning, Reinforcement Learning, Portfolio Optimization, Proximal Policy Optimization.
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	CLASSIFICATION OF MULTISPECTRAL IMAGES FOR LAND USE AND LAND COVER USING TRANSFORMER MODEL
Author:	ADLENE EBENEZER P, S. MANOHAR
Abstract:	Land cover is the term used to describe the geographical coverage that encompasses the planet's surface. It consists of trees, greenery, concrete, barren ground, and water bodies. Remote imaging is generally used to obtain data on the land covering. The information collected is then classified according to different land covers using an approach called land cover classification. In order to predict future changes in land utilisation and land cover for the forest, and non-forest covered regions of the hilly region in the region of Tiruvanamallai, India, this research develops the Vision Transformer framework. With an overall accuracy of 94.01% in 2015 and 94.19% in 2018, the Land Use/Land Cover (LU/LC) classification map during the classification yields positive validation results.
Keywords:	Land Use/Land Cover, Transformer model, Forest, Non-forest, Classification
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	SYSTEM ANALYSIS TO RELIEVE ACCULTURATION STRESS OF INTERNATIONAL STUDENTS: FOCUSING ON KOREAN AND CHINESE WEB PORTAL CASES
Author:	LIQIU SUO1, OOK LEE, CHANGBAE MUM, and HYODONG HA
Abstract:	As the number of foreign students opting for studying abroad continues to rise, there has been a corresponding surge in research and discussion surrounding the process of acculturation for these international students. In line with this trend, it is imperative to assist international students in adapting to study abroad programs and society, while also reducing the stress associated with acculturation. The objective of this research is to ascertain the essential information content required by international students. To accomplish this, a survey was conducted on three Korean websites to evaluate the accuracy of the provided information and the satisfaction levels of students with their learning experiences. The primary research goal of this study is to identify the information content that international students truly require, and to offer a reference and assistance for the future development of a comprehensive acculturation web portal.
Keywords:	Acculturation Pressure, Acculturation, Website Portal, International Students.
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	FERVENT ANT COLONY OPTIMIZATION-BASED DECISION TREES (FACO-DT) FOR ENHANCING CLASSIFICATION ACCURACY IN SENTIMENT ANALYSIS
Author:	G.M.BALAJI, K.VADIVAZHAGAN
Abstract:	Online shopping has transformed how consumers purchase products, offering convenience and a wide range of choices. As customers increasingly rely on online platforms, the role of product reviews has grown significantly. These reviews, submitted by customers who have purchased products online, provide valuable insights to potential buyers about product quality, functionality, and overall satisfaction. However, classifying sentiments accurately from these diverse reviews is a challenging task. Reviews can encompass a broad spectrum of sentiments, often nuanced and context-dependent. Traditional sentiment classification techniques might struggle to handle complex linguistic structures and domain-specific expressions in online product reviews. The proposed work addresses these challenges by introducing the “Fervent Ant Colony Optimization-based Decision Trees (FACO-DT)” approach. FACO-DT harnesses Ant Colony Optimization techniques with Decision Trees to enhance sentiment classification accuracy. Ant Colony Optimization optimizes feature selection, while Decision Trees provide a structured sentiment classification framework. This synergistic approach enables FACO-DT to effectively capture intricate sentiment patterns in reviews, accommodating domain-specific expressions and linguistic complexities. To evaluate the proposed approach, an Amazon product review dataset is utilized. The dataset comprises reviews from Electronics, Industrial, Scientific, and Software domains. Results show that FACO-DT consistently outperforms traditional classification accuracy, recall, and F-measure methods. The approach’s ability to adapt to different domains and enhanced sentiment classification accuracy underscores its potential for real-world sentiment analysis applications. The FACO-DT approach offers a promising solution by fusing optimization and decision-making techniques to handle the complexities of sentiment analysis in diverse online product reviews, as evidenced by its robust performance on the Amazon dataset.
Keywords:	Sentiment, Amazon, Ant Colony, Decision Tree, Optimization
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	A NOVEL APPROACH BASED ON SIMPLIFIED AND IMPROVED FUZZY BWM AND FUZZY TOPSIS FOR GREEN SUPPLIER SELECTION
Author:	EL BETTIOUI WISSAL, ZAIM MOUNIA, SBIHI MOHAMED
Abstract:	Integrating environmentally-friendly practices into industrial processes is crucial to meeting global environmental concerns. This is particularly crucial in the supplier selection process, where environmental considerations intermingle with economic requirements. However, the complexity of this process stems from a number of factors, mainly the uncertainties associated with this decision-making, the lack of comprehensive data and the inherent subjectivity of human judgment. Moreover, Green Supplier Selection requires the participation of multiple decision-makers with different areas and levels of expertise. The methods available to support these complex decisions are often themselves complex, requiring advanced mathematical modeling skills. In order to meet these requirements, this paper proposes a new Green Supplier Selection model combining the use of an improved and simplified Fuzzy BWM and Fuzzy TOPSIS. The key benefit of the suggested model is that it is easy for both researchers and participants to apply in a real-life scenario; it also takes into consideration the subjectivity of human judgment and the diversity of decision-makers. A real-life case study is conducted to demonstrate the effectiveness of our Green Supplier Selection approach. Three green suppliers of a building materials manufacturing company are evaluated. Our approach is validated by comparing its results with two other existing approaches. This comparative study reveals the superiority of the proposed method over previous studies. Thus, our research provides a powerful tool to guide decision-making in the selection of green suppliers.
Keywords:	Group Multi Criteria Decision-Making, Fuzzy Set Theory, Fuzzy BWM, Fuzzy TOPSI, Green Supplier Selection
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	IMPROVING RSSI-BASED DISTANCE PREDICTION BY UTILIZING ROOM TEMPERATURE AND HUMIDITY VARIANCE FOR INDOOR POSITIONING SYSTEM APPLICATION
Author:	DAVID ABRIMAN SIMATUPANG, GEDE PUTRA KUSUMA
Abstract:	The development of cellular telephone technology has increased its function from a means of communication into many functions. One of them is in determining the location and distance inside the room using Bluetooth technology. In this technology, the received signal strength indicator (RSSI) of the Bluetooth radio signal is used to estimate the distance between the signal sender and receiver. Due to the fluctuation of the RSSI, the level of measurement accuracy is low so that the development of pre-processing method is carried out to overcome this problem. The proposed RSSI pre-processing method utilizes room temperature and humidity as environmental parameters processed using the Kalman filter (KF), Support vector regression (SVR), and Multilayer Perceptron (MLP). From the evaluation, it is shown the MLP yield the best result with lowest error and highest accuracy of distance prediction and position compared to other methods. Average of the distance prediction evaluation using MLP utilized 4 BLEs with temperature and humidity evaluation test in high temperature-low humidity. The mean absolute percentage error (MAPE) of the MLP is 5.7% and the mean absolute error (MAE) result is 0.124m. For RSSI-based position prediction test in high temperature and low humidity using MLP, it achieved the mean error (ME) of 0.171m, which is lower than without utilizing temperature and humidity with ME of 0.423m. The RSSI-based distance and position prediction models utilizing temperature and humidity gave lower error and higher accuracy compared to the models that did not use temperature and humidity parameter. By utilizing room temperature and humidity using MLP in our research able to improve the accuracy with lesser error for distance prediction and indoor positioning system (IPS) application compared to KF and SVR method.
Keywords:	Room Temperature and Humidity, RSSI Distance Prediction, Support Vector Regression, Multilayer Perceptron, Indoor Positioning System.
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	MDCGANOCIS: MODIFIED DEEP CONVOLUTIONAL GENERATIVE ADVERSARIAL NETWORKS BASED ORAL CANCER IDENTIFICATION SYSTEM
Author:	DHARANI R , REVATHY S, DANESH K, PEERTHI PARAMESWARI, AISHWARYA B
Abstract:	This paper develops a novel feature extraction model based on Generative Adversarial Networks (GAN) and Convolution Neural Network (CNN) to detect the oral cancer with high accuracy. The main objective of this work is to classify the input Oral Cavity Squamous Cell Carcinoma (OCSCC) image as healthy or sick. The methods used here are, Modified Deep Convolution Generative Adversarial Networks (MDCGAN) as feature extractor and Modified Convolution Neural Network (MCNN) is used for classification oral cancer images. Before extracting process, the first step is to image enhancement. For this step, first input image is resized, contrast enhanced and finally RGB color space is converted into YCbCr color space. For contrast enhancement in this work uses the Improved CLAHE method. The study found that proposed work gives best result than existing approaches. Performance of the proposed technique in terms of classification precision during the testing phase. The suggested method achieving impressive accuracy, precision, recall, and f1-score rates of 97.83%, 97.50%, 95.12%, and 96.30%, respectively, for magnification 400x, and 98.11% accuracy for magnification 100x. during the testing phase. The originality of this work is in the application of the MDCGAN feature extraction model, which is based on deep learning, to obtain pertinent features for classification. The overall quantity and caliber of the features extrapolated from the OCSCC image determine how well oral cancer will be predicted. The accuracy will grow if feature sizes are expanded. GAN is typically used to increase the dataset's image count. However, in our method, deep feature extraction is done via GAN. The generator components of our suggested MDCGAN model operate similarly to conventional GAN. This section is used to increase the dataset's sample size for each image. The accuracy will grow if feature sizes are expanded. However, MCNN replaces the discriminator component of traditional GAN. The detection precision for oral cancer prediction will increase with the use of this innovative MDCGAN feature model. Therefore, MDCGAN is far superior to conventional deep learning algorithms for such image classification applications.
Keywords:	Oral Cancer, Deep Learning, Classifiers, Real-Time, CNN, MDCGAN, DCGAN
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	LOW RESOLUTION FACE RECOGNITION ON CCTV IMAGES USING A COMBINATION OF SUPER RESOLUTION AND FACE RECOGNITION MODELS
Author:	ANTONIUS RILDO PRAMUDYA GONDOSISWOJO, GEDE PUTRA KUSUMA
Abstract:	Closed-Circuit Television (CCTV) serves as an essential device in contemporary society due to its capacity to capture images in public spaces, thereby contributing to the suppression of crime rates. However, a prevalent issue encountered is the small size of the images detected by CCTV, measuring only 32 x 32 pixels, resulting in inadequate facial recognition due to visual blurriness. To address this challenge, researchers opt to enhance the image resolution using the Super Resolution (SR) method before subjecting it to Face Recognition (FR) technology. This combined approach is referred to as Low Resolution Face Recognition (LRFR). In this research, the investigators aim to identify the optimal combination of SR and FR models. The SR models utilized include U-Net, EDSR, and Bicubic Interpolation, while the FR models employed are ResNet50 and MobilenetV2. As a result, six combinations of SR and FR are derived. The dataset employed for this study is LFW (Labelled Faces in The Wild). Based on the evaluation results, the study concludes that the most effective combination of SR and FR models is U-Net and ResNet50, achieving an accuracy rate of 85%, precision of 87%, recall of 85%, and a processing time of 11.454 seconds. Additionally, this combination successfully enhances the image resolution from 32 x 32 pixels to 128 x 128 pixels.
Keywords:	CCTV Images, Super Resolution, Low Resolution, Face Recognition, LFW Dataset
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	A HIERARCHICAL ATTENTION MECHANISM FOR SENSOR DATA ANALYTICS IN INTERNET OF THINGS (IOT) APPLICATIONS
Author:	DR. FRANCISKUS ANTONIUS, R. MANIKANDAN, RICARDO FERNANDO COSIO BORDA, KAMILA IBRAGIMOVA, DILYORJON YULDASHEV, JORGE L. JAVIER VIDALÓN
Abstract:	The rapid proliferation of Internet of Things (IoT) technology has resulted in an exponential increase in sensor data generated by diverse connected devices. Extracting valuable insights from this vast and complex data has become a critical challenge, necessitating advanced analytics techniques. In this project, to improve sensors analysis of data in applications for the Internet of Things, we suggest a unique technique that combines a method of attention with Long Short-Term Memory (LSTM). The attention mechanism selectively focuses on relevant sensors and their readings, dynamically weighting their importance based on the context, allowing intricate trends and connections between dates in the data to be captured by the algorithm. Concurrently, LSTM excels at modeling sequential information, enabling accurate predictions and efficient anomaly detection. Extensive experimentation and performance evaluations are conducted to assess the efficacy of our approach, contrasting it with current practices. The outcomes show that our suggested technique produces improved predictions. accuracy, efficiency, scalability, and robustness to missing data, outperforming other approaches. The synergistic integration of attention mechanism and LSTM empowers IoT applications with deeper insights and more informed decision-making capabilities. This research highlights the potential of advanced analytics techniques in optimizing IoT systems, fostering data-driven innovation, and promoting efficient resource utilization across various industries, including smart manufacturing.
Keywords:	Internet Of Things, Sensor Data Analytics, Attention Mechanism, Hierarchical Attention, Smart Building.
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	ENHANCING SATELLITE IMAGE CLASSIFICATION IN NOISY ENVIRONMENTS WITH DEEP CONVOLUTIONAL NEURAL NETWORKS
Author:	Y.VISHNU TEJ, M. JAMES STEPHEN, P.V.G.D. PRASAD REDDY
Abstract:	Satellite Image Classification is an essential aspect of remote sensing image processing. Satellite imagery has emerged as a pivotal data source for diverse applications such as land cover classification, urban planning, environmental monitoring, and disaster management. Satellite images are often subjected to various sources of noise, which can degrade the performance of traditional image classification techniques. Deep Convolutional Neural Networks (DCNNs) have shown remarkable success in image analysis tasks and have demonstrated potential for handling noisy satellite images. This paper investigates the performance of three popular DCNN architectures, namely VGG-16, ResNet-50, and Inception V4, for classifying noisy satellite images. To create a diverse and challenging dataset, we introduce noise into the original high-resolution satellite images, simulating real-world noise scenarios. The RSI-CB dataset covers various geographic regions and land cover types, encompassing the challenges faced during satellite image analysis. It contains six categories with 33 sub-classes and over 24,000, 256 X 256 pixel images. This paper contributes to advancing the use of DCNNs for satellite image classification in noisy environments. The study offers valuable guidance for selecting appropriate architectures based on the noise characteristics of satellite image datasets, ultimately enhancing the accuracy and reliability of satellite-based applications in challenging real-world conditions.
Keywords:	Deep Convolution Neural Network, Remote Sensing Image Classification, VGG-16, ResNet-50, Inception V4
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	HOW TO INCREASE CUSTOMER TRUST TO PURCHASE GADGETS THROUGH E-COMMERCE PLATFORM
Author:	QUEDI ZATA FAKHRUSY, SFENRIANTO
Abstract:	With the increasing penetration of internet use in Indonesian society, there has been a change in behavior towards daily activities including shopping. Currently, online shopping activities are carried out on e-commerce platforms. On the e-commerce platform, Indonesians can find whatever they need, including gadgets. However, the activity of buying gadgets through e-commerce platforms by Indonesians is less than 28%. The lack of trust from the Indonesian people is one of the reasons Indonesian people are more comfortable buying gadgets at conventional shops. For this reason, researchers want to know the purchasing factors that have a significant impact on customer trust in buying a gadget. This study collected data by distributing questionnaires to respondents who had purchased a gadget on an Indonesian e-commerce platform at DKI Jakarta province. Total respondent data obtained in this study amounted to 156 respondents. This research uses Structural Equation Modeling (SEM) to assess the important factors of customer trust in purchasing gadgets on Indonesia e-commerce such as: seller's reputation and expertise, e-commerce’s reputation, product quality, information quality and perceived guarantee. The results of this study showed e-commerce reputation, quality product and perceived guarantee has significant impact on customer trust in purchasing gadgets on Indonesia e-commerce. The study's findings should give a broad overview of the aspects that influence customer trust while purchasing gadgets on Indonesian e-commerce. And it is hoped that it can provide input to e-commerce platforms (Tokopedia, Shopee, Lazada, BliBli) and customers to pay attention to important factors in buying gadgets on Indonesia e-commerce platforms.
Keywords:	Customer, E-Commerce, Purchasing Gadgets, Structural Equation Modeling (SEM), Trust.
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	KIDNEY STONES IDENTIFICATION UTILIZING EXTREME LEARNING MACHINE
Author:	SHARFINA FAZA, ROMI FADILLAH RAHMAT, MERYATUL HUSNA, AJULIO PADLY SEMBIRING, LISA FELICIA SILAEN, FARHAD NADI
Abstract:	The kidneys are two reddish-brown bean-shaped organs in humans. One of the diseases that attack the kidneys is kidney stones. Kidney stones occur when there are minerals or other substances in the blood that crystallize in the kidneys and form a solid. In identifying kidney stones, doctors and radiologists look at ultrasound images of the kidneys manually. Of course, doctors are experts at detecting them, but there is still the possibility of an error in predicting the images. For this reason, a computational method is needed to facilitate the identification of kidney stones suffered by patients through ultrasound images of the kidneys. This study aimed to identify kidney stones through ultrasound images using the Extreme Learning Machine method. The steps in conducting this research are Preprocessing, Segmentation, Feature Extraction, and Identification. In preprocessing Scaling and contrast enhancement is using CLAHE, in Segmentation uses Thresholding OTSU and Morphological Close segmentation methods, in feature extraction the method used is Gray Level Co-occurrence Matrix (GLCM) and identification with Extreme Machine Learning. This study obtained an accuracy of 92.30%. The contribution of this research is to show the ability of ELM in classifying kidney stone images, which will give initial identification before definite verdict from health specialist.
Keywords:	Kidney, Kidney Stones, Images, Gray Level Co-occurrence Matrix, Extreme Machine Learning
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	ENHANCING TRANSACTION VERIFICATION THROUGH PRUNED MERKLE TREE IN BLOCKCHAIN
Author:	M.GRACY, B. REBECCA JEYAVADHANAM, S. ALBERT ANTONY RAJ
Abstract:	A Merkle tree is a data structure employed within Blockchain technology to securely verify information or transactions within a vast data collection. This paper proposes a new and improved verification method, Pruned Merkle Tree (PMT), for hash nodes marching to the Merkle Root in a Minimal duration. PMT is a unique mechanism for verifying unpaired transactions in a block. The future influence of cryptocurrency will be immense, and PMT showcases its effectiveness in terms of transaction speed and node repetition. Our method allows any block to validate the full availability of transactions without repeating hash nodes and focuses on improving the transaction process through the Pruned Merkle Tree and achieving remarkable results. To assess the performance of the proposed system, we used Hyperledger Caliper, a benchmarking tool specifically designed for measuring the performance of Hyperledger-based blockchain solutions. The evaluation results show a significant improvement in throughput, with a value of 30450kbps recorded. The processing time has also increased noticeably, reaching 1660ms. Security measures have also been strengthened, yielding an impressive 99.60%. The energy consumption factor plays a crucial role, and the PMT exhibits the lowest value at 235 joules.
Keywords:	Blockchain, Merkle Tree, Pruned Merkle Tree, Security, Transaction Verification
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	NEW CRYPTOSYSTEM USING ENHANCED AUTOMATIC THEOREM PROVING AND ENHANCED CIPHERS
Author:	K. KANTHI SOWJANYA, SURENDRA TALARI, GEESALA YOJANA, PRASAD VANGAPANDU
Abstract:	Almost every sector in India has been influenced constructively and positively by globalization. In the present digitalized world, the safety and conduct of information in the cyber space is quite pivotal. In the secure and safe exchange of information, the multitudinous dimension of cryptography plays a major role. Using Automatic Theorem Proving concept, this article proposes a new cryptosystem technique by allocating various enhanced ciphers to the enhanced antecedent and consequent rules. Using plaint text and connective symbols, the sequent is formed; then this plain text is encrypted into various levels using enhanced antecedent rules, consequent rules and the corresponding allocated enhanced ciphers. It involves multiple levels of encryptions and decryptions with more security thus making it harder to the attacker to decipher the plain text. The security levels are infeasible in spite of the feasibility of encryption and decryption run time of the proposed technique.
Keywords:	Sequent, Connectives, Enhanced antecedent rules & consequent rules, Enhanced ciphers, Encryption, Decryption
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	AN INTELLIGENT CYBERSECURITY MODEL FOR IOT NETWORKS USING ADVERSARIALLY REGULARIZED PARALLEL DEEP TRANSFER NETWORK
Author:	LAVANYA VEMULAPALLI, P.CHANDRA SEKHAR
Abstract:	The increasing number of technology and information has led to significant security problems, which have increased the significance of the creation of sophisticated intrusion detection systems (IDS). Big data may be handled through deep learning, which has demonstrated excellent performance in several disciplines. As a result, security professionals want to use deep learning in intrusion detection systems. This subject has been the subject of a great deal of research, which has produced a wide variety of methods. To categorise network traffic, the majority of these methods employ predetermined characteristics that have been retrieved by specialists.However most of the methods are incapable of effectively classifying the intrusion. To solve this problem, this research introduced a novel deep learning method to effectively classify the intrusion with lesser delay. Using a multi-level clustering strategy that includes the k-nearest neighbour, reverse k-nearest neighbour and k-means clustering approach, the undesirable information and outliers in the data may be eliminated. The Adversarially Regularised Parallel Deep Transfer Network (AR-PDTN) is then used to classify the IoT data intrusion in order to identify any network irregularities. The parameters of the proposed neural network model will then be appropriately adjusted using Alpine skiing optimisation. Five independent datasets like CICIDS2017, NSL-KDD, KDDCup99, UNSW-NB15 and BOT-IOT were utilised to categorise the incursion. In the results section, the accuracy, precision, recall, and f1-score of the proposed model are contrasted with those of several other current models.The proposed model achieved 99.3% accuracy in UNSW-NB15 dataset, 99.7% in CICIDS2017, 99.1% in NSL-KDD dataset, 99.8% using BoT-IoT dataset and 94.5% in KDDCup99 dataset, respectively.The performance of the proposed framework was the best when compared with each dataset. In this research, an Adversarially Regularized parallel deep transfer Network (AR-PDTN) is introduced to classify the intrusion from IoT data. The unwanted information and outliers data could be reduced by using multi-level clustering approach concerning k-nearest neighbour, reverse k-nearest neighbour and k-means clustering approach
Keywords:	Linear Time Complexity, Adjacent Neighbours, High Density Points, Wasserstein Auto Encoder, Critic Unit, Adversary Auto Encoder, Down Sampling, Convolutional Filter Tube
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	SUGAR CANE LEAF DISEASE CLASSIFICATION AND IDENTIFICATION USING DEEP MACHINE LEARNING ALGORITHMS
Author:	LAKSHMIKANTH PALETI , ARAVA NAGASRI , P. SUNITHA3 , V SANDYA , T. SUMALLIKA , PRABHAKAR KANDUKURI , K. KISHORE KUMAR
Abstract:	The identification of crop diseases is one of the major concerns that the agricultural industry has to deal with. The detection and classification of leaves is essential in agriculture, forestry, rural medicine, and other commercial applications, among other things. The diagnosis of sugar cane plant leaf disease is required for automatic weed identification in precision agriculture. This paper discusses a novel approach to the development of a plant disease recognition model that is based on sugar cane leaf image classification and employs deep convolutional networks to recognise disease in sugar cane plants. The method used for identification and automatic recognition investigates the possibility of using k-NN and SVM in pre-training with ANN, followed by CNN-based approaches for recognition.
Keywords:	KNN, SVM, Leaf Disease, Calssificaiton, ML
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	TENACIOUS FISH SWARM OPTIMIZATION BASED HIDDEN MARKOV MODEL (TFSO-HMM) FOR AUGMENTED ACCURATE COTTON LEAF DISEASE IDENTIFICATION AND YIELD PREDICTION
Author:	S.GOVINDASAMY, D.JAYARAJ
Abstract:	This research presents an innovative approach called Tenacious Fish Swarm Optimization based Hidden Markov Model (TFSO-HMM) for augmented accurate cotton leaf disease identification and yield prediction. Cotton leaf diseases significantly threaten crop productivity, requiring timely detection and precise prediction for effective disease management. The proposed TFSO-HMM framework combines the strengths of Tenacious Fish Swarm Optimization (TFSO) and the Hidden Markov Model (HMM) to address the challenges associated with disease identification and yield prediction in cotton plants. TFSO, a nature-inspired optimization algorithm, optimizes the classification process, enhancing the accuracy of disease identification. By harnessing the collective intelligence of fish swarms, TFSO intelligently explores the search space to identify the optimal solution. The selected information is then incorporated into the HMM framework, which captures the temporal dependencies in disease progression and yield prediction. HMM's sequential modelling approach facilitates understanding the dynamic behaviour of cotton leaf diseases over time, leading to more accurate predictions. Experimental results on a comprehensive dataset demonstrate the superior performance of the TFSO-HMM method over existing approaches in terms of accuracy and predictive capability. The augmented accuracy achieved through TFSO-HMM enables early detection and precise prediction of cotton leaf diseases, enabling timely interventions for disease management and maximizing crop yield.
Keywords:	Tenacious Fish Swarm Optimization, Hidden Markov Model, Cotton Leaf Disease, Yield Prediction, Disease Identification, Augmented Accuracy.
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	DECISION SUPPORT MODEL FOR DETERMINING CYBERBULLYING TWEET
Author:	DARWIN SAMALO, DITDIT NUGERAHA UTAMA
Abstract:	The study was driven by the difficulty in distinguishing cyberbullying and bullying and realizing the impact if they are not differentiated. The impacts include the increasing number of cyberbullying incidents, the difficulty of teenagers in distinguishing between banter and cyberbullying, and much cyberbullying masked as banter and vice versa. Most researchers focus only on cyberbullying models, unaware that cyberbullying and banter have a fine line that must be distinguished. The proposed DSM-based model can distinguish this using eleven parameters: violence, hate, aggression, swearing terms, dominant personality, emojis, relationship, the severity of harm, imbalance of power, repetition, and visibility among peers. These parameters will be grouped into four categories based on the method used: manual labeling, lexical category, Laplace's rule of succession, and multi-stage fuzzy. As the test results show, only 27 of the 438 cyberbullying datasets are actually cyberbullying. These results prove that the proposed model can distinguish between cyberbullying and banter, in which case the dataset of the previous model was initially classified as cyberbullying but instead was classified as bullying after further analysis.
Keywords:	Decision Support Model, Multi-stage Fuzzy Logic, Cyberbullying and Banter, Classification, Mining Twitter Data
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	THE USE OF DATA MINING FOR MONITORING THE QUALITY OF LIFE OF HEMODIALYSIS PATIENTS
Author:	JAOUAD CHOUIKH, SAMIA RKHA, AIT EL HAJ SALOUA, NEZHA NACER, NADIA OUZENNOU
Abstract:	Data mining is a valuable tool for monitoring the quality of life of hemodialysis patients, a crucial treatment for chronic kidney disease. It involves extracting meaningful information and patterns from large amounts of data, such as electronic medical records, laboratory results, questionnaires, demographic data, and symptom tracking data. Data mining can identify patterns and trends indicating deterioration in quality of life in hemodialysis patients, allowing healthcare professionals to detect early signs of emotional, social, or physical problems, take preventive measures, and adjust treatment accordingly. Historical data mining can predict future quality of life outcomes, allowing healthcare professionals to better understand factors influencing patients' quality of life and develop personalized treatment plans. It can also identify correlations between different symptoms and the quality of life of hemodialysis patients, enabling physicians to tailor treatment protocols to better manage specific symptoms and improve the patient's overall quality of life. However, data mining in the medical field requires an ethical approach and appropriate protection of sensitive patient data. Confidentiality and data security must be strictly adhered to to ensure the responsible use of patient information. This study focuses on monitoring the quality of life of hemodialysis patients using data mining in the Marrakech-Safi region, providing valuable information on treatment effectiveness, factors influencing patient quality of life, and strategies for improving care. Data mining is a valuable tool for monitoring the quality of life of hemodialysis patients. It extracts meaningful information from large amounts of data, such as electronic medical records, laboratory results, questionnaires, demographic data, and symptom tracking data. This process helps identify patterns and trends indicating deterioration in quality of life, allowing healthcare professionals to detect early signs of problems and adjust treatment accordingly. Data mining can also predict future outcomes, helping healthcare professionals develop personalized treatment plans. It can also identify correlations between symptoms and quality of life, enabling physicians to tailor treatment protocols. However, ethical practices and data security must be adhered to. This study focuses on monitoring hemodialysis patients in the Marrakech-Safi region to improve care and management. Data mining is a valuable tool for monitoring the quality of life of hemodialysis patients. It extracts meaningful information from large amounts of data, such as electronic medical records, laboratory results, questionnaires, demographic data, and symptom tracking data. This process helps identify patterns and trends indicating deterioration in quality of life, allowing healthcare professionals to detect early signs of problems and adjust treatment accordingly. Data mining can also predict future outcomes, helping healthcare professionals develop personalized treatment plans. It can also identify correlations between symptoms and quality of life, enabling physicians to tailor treatment protocols. However, ethical practices and data security must be adhered to. This study focuses on monitoring hemodiaysis patients in the Marrakech-Safi region to improve care and management.
Keywords:	Artificial Intelligence , Data Mining , Data Processing , Patient Quality of Life , haemodialysis patients.
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	IMPLEMENTATION OF ELGAMAL AND LEAST SIGNIFICANT BIT (LSB) ALGORITHM FOR ENDING AND HIDDEN MESSAGES IN DIGITAL IMAGES
Author:	TONNI LIMBONG, A M H PARDEDE, DESINTA PURBA, LAMHOT SITORUS
Abstract:	Basically, confidential data needs to be stored or conveyed in a certain way so that it is not known by unauthorized foreign parties. And to overcome this problem, the science of cryptography and steganography was created. Cryptography is the art and science of keeping messages confidential by disguising them in an encoded form that has no meaning, while steganography is the art and science of hiding secret messages inside other messages so that the whereabouts of the secret message cannot be known. Steganography keeps messages secret by hiding messages. The current implementation of steganography uses digital media as a medium for storing or hiding messages, one of which is image media (digital image). The combination of cryptography and steganography can provide better security for secret messages, where secret messages are first encrypted using the ElGamal algorithm, then the ciphertext results from the cryptography are hidden in image media using the Least Significant Bit (LSB) steganography method. The implementation of cryptographic algorithms and steganography methods can further increase the security of secret messages.
Keywords:	Ciphertext, Cryptography, ElGamal, Encryption, Steganography.
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	LOCALIZED NETWORK MODEL FOR IMAGE DENOISING AND PREDICTION USING LEARNING APPROACHES
Author:	A.KARTHIKRAM, M.SARAVANAN
Abstract:	Computer Tomography (CT) imaging provides a promising solution for various health-based evaluations and diagnoses. Certain parametric mappings of cerebral parenchyma are performed with continuous CT scans. It is highly solicited to diminish the CT dosage for constant application owing to the higher radiation exposure due to continuous scans. Thus, there is a need of novel denoising and classification technique. Here, image denoising is essential to attain a reliable diagnosis. This research concentrates on modelling a novel deep learning approach with a Localized convolutional image denoising auto-encoder (L-CNM) for CT image denoising, which avoids the higher-dose referral images during the training process. The proposed network model is trained by mapping the image frames captured from CT and evaluating the adjacent frames. The noise over the CT source is independent, and the proposed model intends to eradicate the noise. The anticipated model can be easily adapted to the various real-time analyses as the model deals effectually with the high-dose training images. The proposed method is validated using the online available public dataset and simulated in MATLAB 2020a environment. The model attains improved image quality compared to various existing denoising approaches.
Keywords:	Computer Tomography, Denoising, Deep Learning, Autoencoder, Training Model
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	MULTI-DIMENSIONAL ASPECT LEVEL HELPFULNESS PREDICTION OF ONLINE CUSTOMER REVIEWS
Author:	ALA A AHMAD KREASHAN, MOHAMMAD SAID EL-BASHIR, ANAS JEBREEN, ATYEH HUSAIN
Abstract:	Estimating and predicting the reviews’ helpfulness become essential for consumers and e-commerce sys-tems that help access the proper reference through massive product reviews. Reviews’ helpfulness is usual-ly calculated based on perceived helpful votes. This study extends the prior review helpfulness studies. It aims to identify review characteristics that best represent the review helpfulness and then use them to pre-dict accurate new helpfulness scores at the minimum possible error. Several natural language processing tools are configured and used to extract review characteristics from the Amazon.com dataset. Six review characteristics (i.e., review age, review aspects, review length, review polarity, review rating, and review subjectivity) that span the three main categories of the review elements were identified as the most influen-tial for review helpfulness and proposed a multiple linear regression (MLR) model that makes use of such characteristics to predict review’s helpfulness. The ability of the proposed model to predict the review help-fulness at minimum error was tested and compared with related prediction methods under various scenari-os. The results show that combining the characteristics associated with the linguistic, content, and peripher-al review elements improves the accuracy of helpfulness prediction, and the proposed MLR model predicts the most accurate helpfulness score at minimum error. The MLR model outperforms the SVM and DT methods by 17.68% and 1.74% in reducing MAE error and by 9.3% and 0.91% in reducing RMSE error, respectively. This study offers a novel contribution to the literature by illustrating the importance of incor-porating the most influential review characteristics in the review helpfulness prediction and how it affects the predictive performance. This study extends ongoing studies on helpfulness prediction and provides notable implications for research and practice; e-commerce systems can have better organization and ranking of their reviews, and customers can efficiently access knowledge to make better purchase deci-sions.
Keywords:	Customer Review, Prediction, Review Helpfulness, Regression Analysis, Multiple Regression.
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	BIG DATA ANALYSIS: THE INFLUENCE OF SCIENCE AND MATHS EDUCATION ON PREGNANCY OUTCOMES: EXAMINING THE MEDIATING EFFECTS OF CUMULATIVE GPA, BIRTH CONTROL, AND BAD HABITS
Author:	TING TIN TIN, CHONG MIN KIT, SAM YEW THENG, NICHOLAS CHAN WEI JIE, SAN WAI CHUNG, TEOH CHONG KEAT, CHAW JUN KIT, LEE KUOK TIUNG
Abstract:	The importance of successful pregnancy outcomes for the necessary fulfilment of different phases of life and ultimately improving families' happiness is often a challenge for the society. Despite many studies that examine the relationship between education level and socioeconomic status, there is a lack of measurement of this relationship in large datasets. This study analyses the relationship between educational attainment in science and maths and pregnancy outcomes, along with the mediating variables cumulative GPA, birth control, and bad habits. 9 variables and a sample size of 1,011 was used to identify how science and maths education influences pregnancy outcomes. A correlation test and mediation analysis were conducted to measure the significant relationship between the variables. This study is significant because it highlights the existence of a positive relationship between education level and the success of women's pregnancy outcomes. Our findings suggest that high levels of education are positively correlated with successful pregnancy outcomes, while possession of bad habits and poor birth control measures had a negative impact on pregnancy outcomes.
Keywords:	Pregnancy outcomes, Science education, Maths education, Cumulative GPA, Birth control, Bad Habits, Mediating effects.
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	EVALUATION OF THE USAGE OF RFID AS IDENTIFICATION FOR STACKED OBJECT DATA RECORDING
Author:	ERVIN SURIANDI, GEDE PUTRA KUSUMA
Abstract:	Radio Frequency Identification (RFID) technology recently has revolutionized inventory management and object tracking. The research begins by utilizing RFID technology to check and process data concurrently for multiple mineral water cup boxes. Furthermore, processing multiple objects that are being read by an RFID reader with various approaches for managing concurrency and parallelism related to data. The result shows the benefit of employing RFID technology in scenarios where multiple objects need to be checked and processed concurrently. The RFID system showcases improved efficiency and scalability compared to traditional methods, allowing it to handle larger volumes of objects with reduced processing time and improved accuracy. This paper shows the number of mineral water cup boxes that could be detected and describe handling data with the utilization of asynchronous mechanisms.
Keywords:	RFID Technology, Object Tracking, Inventory Management, Parallel Computing, Kotlin Coroutines
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	A METHOD TO RANK INTERVAL VALUED TRAPEZOIDAL INTUITIONISTIC FUZZY SETS AND ITS APPLICATION TO ASSIGNMENT PROBLEM
Author:	S. N. MURTY KODUKULLA, SIREESHA VEERAMACHANENI
Abstract:	An Assignment Problem plays a crucial role in industry, decision making analysis and many other applications in engineering and management science. An interval valued trapezoidal intuitionistic fuzzy set (IVTrIFS) is a strong instrument to capture uncertainty. The ranking of IVTrIFSs is a essential whenever fuzzy set theory is applied to study any real-life problem. In this paper, a new method for ranking IVTrIFSs is introduced by using the concept of centre of gravity (COG) of hesitancy degree , which is simple to calculate and easy to apply for comparing IVTrIFSs. The proposed method is compared with the existing methods using numerical examples for its superiority. Further, an assignment problem under IVTrIFSs environment is discussed using the proposed method.
Keywords:	Assignment Problem, Ranking Of Fuzzy Numbers, Interval Valued Intuitionistic Trapezoidal Fuzzy Number (IVITFN), Centre Of Gravity (COG).
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	ENHANCED QUANTUM FEATURE MAP FOR COLOR IMAGE CLASSIFICATION FOR IMPROVED ACCURACY AND COMPUTATIONAL EFFICIENCY
Author:	VENKAT RAMAN B, NAGARATNA P HEGDE
Abstract:	Quantum representation of RGB images and its dimensionality reduction for efficient classification has been still a challenge due to which learning of images with quantum advantage is lagging behind. Although existing quantum image representations and feature maps are slowing coming into practice, it is seen that the existing methods suffer lack of accuracy as the color information is lost in view of dimensionality reduction for efficient classification of images. We present a novel idea in which the image’s color information is preserved and the dimensionality is reduced which can be used for classification with improved accuracy and computational efficiency. The accuracy is compared with the classical artificial neural network and quantum convolutional neural networks which uses existing feature maps.
Keywords:	Feature Map, Quantum Image Representation, Classification, Quantum Classification, Artificial Neural Networks (ANN), Convolutional Neural Networks (CNN)
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	AN EFFICIENT PROGRAMMING SYSTEM FOR OPERATING THEATERS BASED ON DISTRIBUTED ARTIFICIAL INTELLIGENCE
Author:	OUMAIMA HAJJI SOUALFI, ABDERRAHIM HAJJI SOUALFI , ABDELLAH EL BARKANY, BILAL HARRAS
Abstract:	The operating theater is an essential element of any hospital structure. It is distinguished by the heterogeneity of its components, some of which have antagonistic characteristics, and receives emergency, planned or ambulatory surgery on a permanent and unpredictable basis. The management of an operating theater involves multiple human and material elements to ensure better quality of medical services for the benefit of patients. In this study, we propose a planning tool that uses a multi-agent planner based on distributed artificial intelligence to ensure efficient and reactive surgical planning that satisfies the various demands of surgical teams despite the constraints encountered, contributing to maximizing operating room occupancy and thus improving operating theater performance and patient satisfaction. We describe the multi-agent system in relation to surgical planning, before presenting its application on a real case of a surgical department. The structure of the tool is detailed, along with its interaction mechanism and decision logic for each step of the planning procedure. The aim of our study is to create a dynamic and weekly operating theater program, taking into account the various incidents that can alter the normal sequence of surgical procedures, and responding to the increased needs of surgical teams and patients.
Keywords:	Multi Agent Planner, Distributed Artificial Intelligence, Planning, Scheduling, Operating Theater.
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	UYAKUY - BLOCKCHAIN-BASED SOFTWARE APPLICATION FOR CERTIFICATE ISSUANCE: A Case Study in the IEEE Peru Section
Author:	JAVIER GAMBOA-CRUZADO , LUIS A. MENDOZA-CHATE , ADONIS VENTOCILLA-SANCHEZ , JEFFERSON LÓPEZ-GOYCOCHEA, JIMMY RAMIREZ VILLACORTA
Abstract:	Currently, companies in the financial sector are widely adopting blockchain technology to address critical issues such as the verifiability of authentic documents and money laundering, among others. However, there remains a prevalence of illicit activities related to document forgery, which is a punishable offense carrying prison sentences. Document verification is a complex process, involving a myriad of challenges and burdens concerning authentication. In this paper, we present an optimized approach for certificate issuance through a blockchain-based web system, with the aim of enhancing certificate authenticity, reducing execution time, and augmenting customer satisfaction. Recently, blockchain has emerged as a promising tool for authenticating the document verification process and is viewed as a significant instrument in combating fraud and document misuse. This research zeroes in on leveraging decentralized technology to create verifiable documents, employing a specific architecture, and following the Rational Unified Process methodology. The results garnered illustrate an efficient process for issuing highly reliable certificates, thereby contributing to the deterrence of malicious use of certificates by third parties.
Keywords:	Web-based system, Blockchain, Issuance, Certificates, Documents, Technology, Dapps
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	THE EFFECT OF TEACHING BY USING COMPUTERIZED ACTIVITIES ON KINDERGARTEN STUDENTS ACQUISITION OF SOCIAL VALUES
Author:	NASEER AHMAD ALKHAWALDEH, HUSSAM N. FAKHOURI, AMANI MATAR
Abstract:	This research paper examines the effect of teaching through computerized activities on the acquisition of social values among kindergarten students. The study employs a semi-experimental design, with an experimental group taught using computerized activities and a control group taught through traditional methods. The research tool used is a social values scale, which measures creative thinking values, communication values, and problem-solving values. The data collected from pre- and post-measurements of social values are analyzed using statistical techniques such as MANCOVA and ANCOVA. The results indicate that teaching through computerized activities has a significant positive impact on the development of social values among kindergarten students. The experimental group showed higher average scores and improved performance in the domains of social values compared to the control group. These findings align with previous research on the benefits of technology integration and computer-assisted instruction in promoting social-emotional development and pro-social behaviors. However, further research is needed to explore the long-term effects and sustainability of computerized activities in early childhood education. The results of this study contribute to the growing body of knowledge on the importance of integrating technology into kindergarten education to enhance social values acquisition.
Keywords:	Computerized Activities, Social Values, Children's School Children
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text

Title:	UNVEILING THE POTENTIAL OF MIXED REALITY: OPPORTUNITIES, CHALLENGES AND FUTURE PROSPECTS
Author:	CHE KU NURAINI CHE KU MOHD, FAAIZAH SHAHBODIN, MULIATI SEDEK, TUAN NORHAFIZAH TUAN ZAKARIA, ANTHONY ANGGRAWAN, HELMI MOHD KASIM
Abstract:	Mixed Reality (MR) is an advancing technology with the potential to revolutionize the realm of information and communication technology (ICT). By combining the strengths of virtual reality (VR) and augmented reality (AR), MR creates an immersive and interactive user experience. Its applications in ICT are extensive, encompassing education, training, entertainment, and healthcare. However, MR is a relatively new technology and faces numerous challenges that must be addressed before its full realization. Many technical aspects and requirements remain unclear. This paper aims to emphasize the applications of mixed reality in the ICT industry. It provides an overview of MR in education, explores opportunities, challenges, and future prospects. Additionally, the authors discuss the opportunities and challenges in MR development. MR technology is widely utilized in art and design, particularly in product design, display design, and interactive design. The paper concludes by summarizing the future trends of mixed reality technology, considering market and industry factors. While the application of mixed reality technology in the field of art design gains increasing attention from industry, it still lacks sufficient attention from the academic community.
Keywords:	Augmented Reality (AR), Immersive Technology, Mixed Reality (MR), Virtual Reality (VR), User Experience
Source:	Journal of Theoretical and Applied Information Technology 31^st October 2023 -- Vol. 101. No. 20-- 2023
Full Text