		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

September 2023 | Vol. 101 No.18

Title:	PREDICTION OF SENSOR DEVICES FAILURE IN UNMANNED AERIAL VEHICLES USING KALMAN FILTER & PARTICLE FILTER
Author:	JALAJAKSHI V, DR. MYNA A N
Abstract:	Unmanned aerial vehicles like drones or certain type of helicopters are having rising importance over the past few decades due to utilizing them in various applications. Since there is no human pilot to respond to any aberrant event, reliability is an important concern in UAV operation. The numerous sensing systems and aerospace navigational instruments used in large aero planes is not applicable to be installed in small UAVs because of the size and budget limitations. As a result, various measures such as analysis redundancies should indeed be used to detect and boost the dependability of positioning instruments. Based on cognitive redundancies, this work provides a sensory fault diagnosis and diagnostic system for tiny unmanned drones. The defect is detected by comparing any significant change in the aircraft’s behavior to the mistake performance, which is approximated with the help of an eyewitness. With the Extended Kalman filter Identification, the observation is derived from input-output observational data. Utilizing feedback experimental observations, the Kalman methodology may recognize the systems and an observers with qualities comparable to a Kalman filter and also provide an extra care with incorporating particle filter mechanism to minimize the noise in the equipment to the least value possible. The suggested technique yields comparable findings to the Kalman filter and in addition to particle filter, and there’s no need to calculate systems matrix or sensors and processing noise covariance. The technology was solely tested with actual drones aircraft performance, and the findings were matched to those obtained using alternative methods.
Keywords:	Drone , Kalman Filter, UAV , Prediction , Particle Filter
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	UNSUPERVISED LEARNING ONTOLOGY BASE TEXT SUMMARIZATIONS APPROACH WITH CELLULAR LEARNING AUTOMATA
Author:	ELHAM GHASEMI, VAHID RAFIEI, GOLSHID RANJBARAN
Abstract:	Text summarization is the procedure for generating a short copy of a given text. The main objective of text summarization is to create an outline that encompasses the text’s main content. In this paper, a new model based on ontology, unsupervised learning, and cellular learning automata is proposed for the text summarization task. For this purpose, using the ontology, concepts of sentences are extracted and mapped to some clusters of sentences with similar meaning, where the most appropriate sentence is selected for the summarization task. The clustering has been done by using K-means unsupervised learning on a corpus of English sentences. Cellular Learning Automata (CLA) is applied for the calculation of n-grams and extracting the summary content. The results were evaluated using the ROUGE-2 method and showed that the quality of the summary text improved by an average of 19.26% compared to other works in the literature.
Keywords:	Text Summarizations, Unsupervised Learning, CLA, Ontology
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	COMPARISON OF STOUT CODE AND FIBONACCI CODE ALGORITHM FOR FILE COMPRESSION BASED ON ANDROID
Author:	HANDRIZAL, T. HENNY FEBRIANA HARUMY, FADHLI IBRAHIM SIREGAR
Abstract:	In the current era, human activities are closely intertwined with the utilization of computers and the internet. Engaging in activities with computers necessitates data storage, whether through cloud-based systems or physical storage mediums. The size of data becomes a critical factor in optimizing resource efficiency for these activities. As a result, ongoing research focuses on data size reduction techniques to enhance overall efficiency. Data compression is the process of converting data into smaller sizes. There are various data compression algorithms. Stout Code algorithm and Fibonacci Code algorithm will be used in this study. We built an Android application and performed a text file compression test using these two algorithms to compare their performance. The comparison parameters that will be used are compression ratio, compression time, and decompression time. The test results indicate that the Stout Code algorithm outperforms the Fibonacci Code algorithm in terms of compression ratio for both homogeneous and heterogeneous strings. The average compression ratios for the Stout Code are 1.949 and 1.159, while for the Fibonacci Code, they are 1.943 and 1.064, respectively. However, concerning compression time and decompression time, the Fibonacci Code algorithm proves to be more efficient. Its average compression times for homogeneous strings and heterogeneous strings are 2437 ms and 2855.429 ms, whereas the Stout Code algorithm takes an average of 2564.857 ms and 3021.571 ms. Similarly, for decompression time, the Fibonacci Code algorithm outperforms the Stout Code algorithm with average times of 349.571 ms for homogeneous strings and 853.857 ms for heterogeneous strings, while the Stout Code algorithm shows average times of 456 ms and 1016.143 ms, respectively. The results lead to the conclusion that the Stout code algorithm outperforms in reducing file sizes, whereas the Fibonacci code algorithm excels in terms of speed.
Keywords:	Stout Code, Fibonacci, Android, Algorithm, Compression
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	IDENTIFYING DIFFERENT EMOTIONS OF HUMAN USING EEG SIGNALS USING DEEP LEARNING TECHNIQUES
Author:	RAJESWARI RAJESH IMMANUEL ,S.K.B.SANGEETHA
Abstract:	Emotions encompass a wide range of feelings, thoughts, and behaviors, reflecting the complex output of the human brain. This interdisciplinary field draws from computer science, AI, neurology, healthcare, and more to study emotional experiences. Understanding and labeling one's emotions are crucial for mental health and well-being, especially in managing stress-related conditions. Emotion classification using electroencephalogram (EEG) signals has gained interest, particularly in affective computing.Developing an effective brain-computer interface (BCI) system for emotion recognition through EEG involves key components such as feature extraction and classifier selection. Deep learning methods, known for their superior performance, have recently garnered significant attention in this domain. Our paper introduces the Deep CNN for Emotion Recognition (DCNNER) framework, utilizing deep convolutional neural networks to accurately detect human emotions from EEG signals.To enhance the model's efficiency, we employ principal component analysis (PCA) for feature extraction(FE) and dimensionality reduction. By feeding only the selected features to various classifiers, we compare their performances on pre-processed and PCA-applied data. The proposed system outperforms existing approaches, achieving a remarkable model accuracy of 99% and a model loss of 0.3. The model employs a 3-dimensional representation, encompassing valence, arousal, and dominance for emotion detection.Our research showcases the potential of deep learning in EEG-based emotion recognition, promising advancements in affective computing and its applications in various domains, including mental health and well-being.
Keywords:	EEG Signal, Stress, CNN, Deep Learning, PCA, Emotion Recognition, Accuracy
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	IMPROVING CASH FORECASTING PERFORMANCE USING A NEW HYBRID-ANN-ARIMA-EM MODEL
Author:	ISAM AHMED M. YAQOOB, KHAIRUL AZHAR KASMIRAN, TEH NORANIS MOHD ARIS, NOR AZURA HUSIN, MOHD YUNUS SHARUM
Abstract:	Budgeting is the technique and science of dividing available and obtainable money between competing needs. Government spending assists programs in tooling up a wide range of services to numerous different slices of the population. Therefore, the demands for further and better services commonly might exceed the government’s capability to provide payments for them. Forecasting of cash flow is the most important instrument for any business in general and exclusively in public budgeting. In simple words, forecasting can show if your business will have sufficient and enough cash to execute the business and/or the ability to expand it. The COVID-19 pandemic has caused an economic shock all over the world. The effect of some external factors such as the COVID-19 pandemic is one of the most important drawbacks that make statistical calculations difficult and sometimes intractable because this external factor is unmeasurable. Artificial intelligence and machine learning have been used widely to improve cash forecasting in many circumstances. In this paper, a comparison between ARIMA, ANN, and hybrid models will be done as a first step to highlight their pros and cons. In the second step, the external factor will be measured to assign the weight. Finally, the new proposed model (Hybrid-ARIMA-ANN-EM) will be explained and implemented, thereafter, by using RMSE and MAPE techniques, differentiation will be applied between the latter and other nominated models revealing the variance in forecasting results between them in terms of accuracy.
Keywords:	ARIMA, ANN, Machine Learning, Artificial Neural Network, Cash Forecasting.
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	TOWARD IMPROVING STUDENT PERFORMANCE IN AN ONLINE ENVIRONMENT USING EDUCATIONAL DATA MINING APPROACH
Author:	ABDULKREAM A. ALSULAMI, , ABDULLAH S. AL-MALAISE AL-GHAMDI, MAHMOUD RAGAB
Abstract:	As online learning grows increasingly popular; there is an increasing demand for practical strategies for enhancing student performance. Analyzing Student performance is one of the most significant issues for decision-makers. Educational data mining techniques are useful for exploring hidden data inside data, finding a pattern, and analyzing student performance. In this paper, we present a hybrid methodology for improving prediction and optimizing student learning outcomes by combining educational data mining (EDM) techniques with ensemble methods. We performed experiments with an online dataset to compare the performance of our proposed model to classic EDM and ensemble approaches. The results demonstrate that our model outperformed the other techniques and obtained a considerable increase in accuracy. Our findings imply that integrating EDM approaches with ensemble methods can increase student performance in online learning environments. This research has significant implications for educators and researchers who want to improve the performance of students and optimize the use of data-driven approaches in online education.
Keywords:	Student Performance; Data Mining; E-Learning; Ensemble Techniques
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	TEAM BUILDING BASED ON A RELATIONAL ANALYSIS ON A PROFILING SYSTEM
Author:	RAFAE ABDERRAHIM, ERRITALI MOHAMMED
Abstract:	In this article, we will show the application of machine learning techniques to develop a recommendation system for employee profiles in order to assign them to specific projects. The aim of this system is to help managers to have a clear idea about the profiles of their collaborators in order to build up a productive team. Our work presents a recommendation system based on a profiling system and relational profile which is based on the publications posted in a professional network. It is made up of two parts. The first part is dedicated to the profiling system based on the extraction of information from internal data and the extraction of interests, psychological profile, and relational profile from the publications exchanged in the professional platform, and the second part is dedicated to the recommendation of profiles corresponds to the requirements of each project.
Keywords:	Recommendation System, Centers Of Interest, Psychological Profile, Relational Profile.
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	REINFORCEMENT LEARNING BASED LOAD BALANCING FOR FOG-CLOUD COMPUTING SYSTEMS: AN OPTIMIZATION APPROACH
Author:	MUSTAFA AL-HASHIMI, AMIR RIZAAN RAHIMAN2, ABDULLAH MUHAMMED, NOR ASILAH WATI HAMID
Abstract:	Fog-cloud computing is a promising approach to enhance distributed systems’ efficiency and performance. Though, managing resources and balancing workloads in such environments remains challenging due to their inherent complexity and dynamic nature. The need for effective load-balancing techniques in fog-cloud computing systems is crucial to optimize resource allocation, minimize delays, and maximize throughput. This article presents a reinforcement learning (RL)-based load balancing system for fog-cloud computing, employing two RL agents: one for allocating new tasks to fog or cloud nodes and another for migrating tasks between fog nodes to ensure fair distribution and increased throughput. This study derived up with novel state, action, and reward models for both agents, facilitating collaboration during the load-balancing process. Three types of rewards for the RL agents are explored: single objective, multi-objective under non-dominated sorting, and multi-objective under lexicographical sorting. The performance of these methods is assessed using metrics such as average utilization, number of tasks completed, serve rate, and delay. The experimental results showed that RL-based scheduling methods, particularly the Reinforce Learning Multiple Objective (RLRLM) with RL-based migration method outperforms greedy on CPU (GR_c) and greedy on reliability (GR_r) methods across all performance metrics. The choice of migration method and reward type also influences performance. These finding highlight RL’s potential in optimizing fog-cloud computing and offer valuable insights for future research and practical applications in this field.
Keywords:	Fog-Cloud Computing, Load Balancing, Reinforcement Learning, Resource Allocation, Multi-Objective Optimization.
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	MULTI-CONTROLLER SDN FOR WIRELESS HETEROGENEOUS NETOWORKS
Author:	JUNHYUK PARK, WONYONG YOON
Abstract:	The state-of-art network has become very complex and heterogeneous to accommodate explosively increasing traffic. In a wireless heterogeneous network, users experience handover between heterogeneous radio access technologies. By separating the control plane from the data forwarding plane, software-defined networking (SDN) enables the flexible management of heterogeneous wireless network resources. However, the performance of a single-controller SDN is limited and difficult to manage with flexibility and agility. The deployment of multi-controllers is more desirable in this kind of networks to improve the scalability and reliability of the control plane. This paper proposes a novel network architecture with loosely-coupled multi-controllers. The loosely-coupled method applied with a multi-controller architecture brings two benefits: one to reduce overburdened control messages to the controller and the other to assure fast response time. Through numerical analysis, we demonstrate that our proposed multi-controller architecture reduces handover cost by 28% and handover delay by 23% compared to the traditional structure. Our architecture also exhibits superior performance compared to other prior architectures in terms of both handover cost and delay, and thereby the proposed multi-controller architecture being an efficient solution for managing the ever-increasing network traffic.
Keywords:	Software-Defined Networking, Multi-Controller, Heterogeneous Network, Loosely-Coupled Network, Handover
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	ENHANCING DATA TRANSMISSION FOR INTELLIGENT INFORMATION SYSTEMS USING SDN TECHNIQUE AND MACHINE LEARNING ALGORITHM
Author:	MOHAMED SALEH YADAM, HAZEM EL-BAKRY, SAMIR ABDELRAZEK
Abstract:	Intelligent information systems, which have recently undergone development and complexity, are now indispensable to the entire world. The networking strategy has unquestionably altered based on machine learning principles to be programable and dynamically configurable with the greatest flexibility and simplicity of use. The term "software-defined network" (SDN) refers to networks that are managed using software applications and SDN controllers as opposed to the more traditional network management consoles and commands, which require a lot of administrative overhead. To centralize network control and administration, SDN changed the topology of network devices to be more flexible and programable. The software-defined network's uses protocols for interacting with and managing switches is called OpenFlow (OF). With this protocol, the switches learn the routing information from the controller and then pass data packets based on this information. One of the most important components of the SDN is the controller, which is the smartest component of the network such as the Ryu controller. Including the importance of the Ryu controller in SDN. This article discussed how to enhance data traffic transmission and classification in the SDN environment. This research shows how we can track all data packets and traffics and automatically identify all data types and classify them correctly, so we can apply a security policy, bandwidth, and quota for each type. The most different thing we used is using a real SDN network environment and also connected a real physical lambda server that makes daily continuous training for all data traffic and synchs this at the same time with the SDN controller that applies this instantly on the real live traffic. Using Machine Learning (ML) and Artificial Intelligence (AI) to enhance the SDN environments and identify data traffic types automatically. The controller (using ML and AI) takes the needed action automatically according to the data types. Enhance security, Data Transmission, and Data Availability in the software-defined networking and Intelligent Systems environment.
Keywords:	Software-Defined Networking (SDN), Machine Learning, Information Systems, Artificial Intelligence, OpenFlow
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	AN ENSEMBLE MOVIE RECOMMENDER SYSTEM BASED ON STACKING
Author:	NISHA SHARMA, DR. MALA DUTTA
Abstract:	In the digital realm, recommender systems are information-filtering algorithms that influence consumer behavior by recommending appropriate products to users. We have proposed an ensemble movie recommender system based on the stacking technique of ensemble learning. The MovieLens dataset from the Grouplens project was used to evaluate the proposed recommender system. The primary objective of this study is to assess the effectiveness of an ensemble movie recommender system based on the stacking technique of ensemble learning. By exploring both standalone and layered models, we aim to demonstrate the potential benefits of ensemble learning in the context of movie recommendations. We have tested with a variety of base learner combinations to identify the ideal base learner and meta-learner pairing for the proposed recommender system. Regression models, including K-Nearest Neighbors and Linear Regressor, and ensemble models such as Random Forest, GradientBoost, XGBoost (eXtreme Gradient Boosting), and AdaBoost have all been tested for this purpose. According to the experiments, layered models perform better than standalone models. Additionally, it has been found that XGBoost performs exceptionally well both as a base learner and as a meta-learner in the proposed stacked model.The main aim of this paper is to increase the model's overall performance and comparison with existing works. We have tested our model using evaluation measures including mean absolute error (MAE), mean squared error (MSE), and root mean square error (RMSE). Our proposed model with 0.69 MAE, 0.81 MSE, and 0.90 RMSE is better than existing works with (0.78 MAE) [17], (0.75 MAE) [18], and (0.70 MAE, 0.83 MSE, 0.91 RMSE) [19].
Keywords:	Ensemble learning, Recommender system, Stacking, XGBoost, Classification, Regression
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	A CRITICAL SUCCESS FACTORS MODEL FOR ADOPTING AND IMPLEMENTING SMART CONTRACTS
Author:	MOHAMED O. GRIDA, SAMAH ABD ELRAHMAN, KHALID A. ELDRANDALY
Abstract:	Recently, smart contracts (SCs) have flourished and have become a mainstream research topic because smart contracts drive the new wave of innovation in business processes. SCs transform real-world contract terms into digital promises of the virtual world. But, the adoption of SCs' technologies in various industries and services is a challenging task. Therefore, this study articulated twenty-one critical success factors for SCs adoption in different applications based on the previous researches and classified them into three categories. These factors are analyzed for each stage of the SCs' lifecycle using the Hierarchical Decision-Making Trial and Evaluation Laboratory (Hierarchical-DEMATEL) technique. At the early stage of SC creation, mature technology, and complexity are the top critical success factors. Then, at the deployment stage, the infrastructural facility and the correctness and immutability of the contract represent the success cornerstones. Finally, having the appropriate infrastructural facility is vital to execute SCs successfully. Based on these results, the authors proposed a smart contract adoption success model based on the implementation life cycle of SCs. The proposed success model was validated by creating a smart contract platform for Egypt's subsidized social housing land program. As a result, the framework managed the project successfully through its early life stages.
Keywords:	Subsidy programs, Critical Success Factors (CSF), Hierarchical DEMATEL, Smart Contract Lifecycle.
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	ENHANCING QUALITY OF SERVICE IN INTERNET OF THINGS-BASED CLOUD WIRELESS SENSOR NETWORKS (IC-WSN) USING ROBUST FROG LEAP INSPIRED ROUTING PROTOCOL (RFLIRP)
Author:	J.JERLIN ADAIKALA SUNDARI, G.PREETHI
Abstract:	Efficient energy management in routing algorithms for Internet of Things-based Cloud Wireless Sensor Networks (IC-WSN) is crucial for greenhouse farming applications. This paper addresses the problem of energy efficiency in routing algorithms for IC-WSN in greenhouse farming. The reliance on battery-powered wireless sensors in greenhouses necessitates the development of innovative routing protocols that can extend sensor lifespan and minimize maintenance costs. In large-scale greenhouse environments with high sensor density, inefficient routing protocols can lead to excessive energy consumption and premature battery depletion. The Robust Frog Leap Inspired Routing Protocol (RFLIRP) is proposed to address this challenge. RFLIRP intelligently selects energy-efficient paths, considering individual sensor energy levels, distance to the destination, and available alternative routes. By incorporating techniques like data aggregation and compression, RFLIRP significantly reduces energy consumption and enhances the operational life of sensors. This research aims to promote sustainability in agricultural practices by optimizing energy consumption, minimizing maintenance costs, and facilitating uninterrupted data transmission for timely monitoring in greenhouse farming. The results highlight the significance of RFLIRP in improving energy efficiency and operational effectiveness in IC-WSN routing for greenhouse farming, paving the way for sustainable and optimized agricultural practices.
Keywords:	Agricultural, Efficiency, Farming, Greenhouse, Routing, Wireless
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	BIG DATA-DRIVEN SUPPORT SYSTEM FOR YOUTUBE CHANNEL IMPROVEMENT
Author:	K. SUBHA, N. BHARATHI
Abstract:	YouTube is gaining a lot of traction and popularity. It has the potential to affect billions of people around the world, as the number of YouTube users continues to rise. YouTube is a video streaming platform owned by Google, with billions of subscribers and 400 hours of video posted every minute. Large volumes with complex data are called big data. The social network YouTube is one of the sources for generating such a high volume of data called big data. YouTube data is not structured data. It is a big challenge to store, process, and analyze such big data in real-time. YouTubers can check their channel performance with YouTube Analytics. The problem with YouTube Analytics is that it's impossible to check another competitor’s channel. The proposed system will analyze real-time YouTube data from the list of channels. It will assist the YouTuber in finding out the Competitor’s channel and how the competitors are doing well on social media platforms such as YouTube. The proposed work analyzes the YouTube data, finding competitors' channels using novel algorithms, and the results are represented in graphical form, which can be utilized by the person or any organization for their decision to improve their revenue.
Keywords:	YouTube, Data analysis, Big Data, Channel statistics, Real-time data.
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	ACF-GSVM: CASCADE AGGREGATE CHANNEL FEATURE WITH GABOR FILTERS AND SUPPORT VECTOR MACHINE FOR ENHANCED FACE DETECTION
Author:	AGIELA OMAR MOHAMED ALGASSIR, SELVAKUMAR MANICKAM, MOHAMMED ANBAR, ASAMA KUDER NSEAF
Abstract:	Face detection has recently attracted a lot of attention because of the groundbreaking work of Viola and Jones. Despite the efforts to improve the performance using several subsequent enhancements, feature representation for face detection remains unsuitable for handling faces with a complicated appearance in an efficient and effective manner. In order to solve this dilemma, we look into the concept of face detection that elaborates the channel overview in a variety of ways, such as using gradient, oriented gradient, histograms, which are known to be efficient for decoding uncomplicated sound information. Consequently, the researchers suggest a novel hybridized and improved face detection method that addresses the problems associated with face detection schemes by utilizing an aggregate channel feature with Gabor filters and a support vector machine, called the ACF-GSVM model. The procedure for improving the face detection is based on a cascade of an aggregate channel feature with Gabor filters and a support vector machine, where the bounding boxes of the face region are initially detected with the use of aggregate channel features. The key decision is made by using Gabor filters and a support vector machine that identifies which bounding boxes belong to the face region. Experiments on the CASIA-FaceV5 database with the proposed ACF-GSVM face detection scheme showed that the accuracy rate it achieved in face detection was 96.6%. This study demonstrates the effectiveness of the proposed method.
Keywords:	Face Detection; ACF; Gabor Filters; SVM; CASIA-FaceV5; Decision Model
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	AN INVESTIGATION INTO THE ROLE OF ONLINE REVIEW FACTORS AND WEBSITE REPUTATION ON CONSUMER PURCHASING BEHAVIOR TOWARDS ELECTRONIC PRODUCTS
Author:	AHMAD SHWEKEH, MOHANNAD MOUFEED AYYASH
Abstract:	Within the realm of current research, a noticeable gap has been identified pertaining to the examination of the impact of online reviews on consumer purchasing behavior, particularly within the context of developing countries. This study uses the Stimulus-Organization-Response (S-O-R) model to evaluate how online review factors affect Consumer purchasing behavior towards electronic products in developing countries. Using 190 customers, review volume, valence, quality, and website reputation were examined on customer buy behavior intentions. The Statistical Package for Social Sciences (SPSS) 26 was used to analyze the data for this study. Results revealed that review quality and website reputation positively and significantly affect consumer purchasing behavior intention, emphasizing the importance of well-written, informative, trustworthy online reviews and a strong website reputation to foster consumer trust and confidence. In contrast, review volume was found to significantly and negatively impact consumer purchasing behavior intentions, suggesting that consumers may be more sensitive to negative information in reviews and overwhelmed by a large volume of reviews, consequently reducing purchasing intentions. In addition, the result revealed a non-significant effect of review valence and consumer purchasing behavior intention. These findings contribute to understanding how various online review factors influence consumer behavior in developing countries with lower electronic product adoption rates and provide insights for practitioners, managers, and researchers to develop targeted strategies that focus on enhancing online review factors and website reputation.
Keywords:	Online Review, Stimulus-Organism-Response (SOR) Model, Reviews Quality, Review Valence, Review Volume, Website, Reputation, Purchasing Behavior, Electronic Products
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	SMART ENERGY EFFICIENT TECHNIQUES FOR IoT ENABLED WIRELESS NODE
Author:	SANGAM MALLA, PRABHAT KUMAR SAHU, SRIKANTA PATNAIK, MANJUSHREE NAYAK
Abstract:	Renewable power transitions as well as decreasing local weather shift need the integration of inexhaustible vitality as well as power use. The IoT along with other contemporary solutions have a broad range of purposes within the power business which includes electricity generation, transmission, distribution and then need. The IoT is usually utilized to improve electricity effectiveness, enhance inexhaustible energy consumption and lower green impacts of energy consumption. The present literature on IoT found smart grids and energy systems is assessed within these specific papers. In this paper, we talk about the allowing solutions of IoT, like different energy efficient devices and cloud computing for information evaluation. We likewise analyze the secrecy as well as protection worries connected with the usage of IoT within the electricity segment, plus recommend a few remedies like sleep mode technologies with many energy harvesting techniques using energy efficient protocols and also increasing the battery life. This particular survey offers an introduction to the job of IoT within improving electricity effectiveness for policy makers, managers and energy analysts.
Keywords:	Internet of Things (IoT), Smart Security System, Smart Sensors, Energy Generation, Wireless Protocols, Sleep Mode, Energy Harvesting Techniques.
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	AN ENSEMBLE STACKING MODEL FOR RUMOR DETECTION BASED ON ARABIC TWEETS
Author:	THANAA MOHAMED HASSAN, YEHIA MOSTAFA HELMY, DOAA S. ELZANFALY
Abstract:	Effective rumor detection within Arabic-language social networks is crucial for mitigating misinformation's impact on users. Despite extensive research on rumor detection in English, investigations into Arabic rumors are limited, despite Arabic's significance for 25 nations. This study proposes a unique hybrid ensemble model for detecting Arabic rumor tweets, addressing both Ensemble stacking-based Machine Learning and Deep Learning dimensions. The initial phase of the proposed model involves the strategic implementation of Machine Learning stacking, wherein standalone classifiers, including the Decision Tree, Random Forest, and Gaussian Naive Bayesian models, are thoughtfully employed. Impressively, the Random Forest and Gaussian Naive Bayesian models exhibit commendable accuracies, both attaining 86%, while the Decision Tree model registers a respectable accuracy of 84%. To further amplify accuracy, a subsequent stage incorporates logistic regression, culminating in an overall accuracy of 87%. In the pursuit of advancing the bounds of accuracy and performance, the study extends its exploration to the realm of deep learning stacking. This facet is manifested through the construction of four distinct neural network models, the collective accuracy of which varies between the noteworthy ranges of 77% to 89%. By judiciously integrating these neural network models with logistic regression in the ensuing stage, the accuracy remarkably ascends to an impressive 90%. This enhancement, amounting to a 3% increase over the machine learning stacking approach, substantially augments vital performance metrics encompassing precision, accuracy, recall, and F1-Score, thus significantly refining the landscape of rumor detection.
Keywords:	Rumor Detection; Online Social Media Networks; Machine Learning; Deep Learning; Ensemble Stacking Model
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	COMMON PRACTICES OF TEACHING AND LEARNING IN REQUIREMENT ENGINEERING SUBJECT
Author:	NURAZLINA MD SANUSI, ASMA ASDAYANA IBRAHIM, MASSILA KAMALRUDIN
Abstract:	Students learn ability in requirement engineering education is very important to make sure an effective teaching and learning process in education field. An effective student learn ability in requirement engineering education is crucial to make sure that the process of teaching and learning is a successful. A successful teaching and learning will produce a proactive, creative, and innovative student in the future. In term of requirement engineering education A successful teaching and learning will produce a proactive, creative and innovative fresh graduate and to be software engineer who worth working in the world wide industry market in the software engineering field. The problem and issues in the industry regarding the software engineering fresh graduates such as poor communication skills, leadership style, ego, gender issue, poor documentation skills, misinterpretation of requirements, incorrect requirements and etc. will no longer become a mess to us if we can apply the new approach in the learning and teaching in the requirement engineering field. This paper reports on a study that identify approach that can help in teaching and learning that focused on student engagement learning. The survey consists of a series of questions about the ability of understanding in requirement engineering activities. The answer to these questions helps to understand common practices, whether the research on pair work in requirement engineering activities and techniques.
Keywords:	Requirement Engineering, Teaching And Learning, RE Subject, RE Education
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	DETECTION OF SARCASTIC SENTIMENT ANALYSIS IN TWEETS USING LSTM WITH IMPROVED ATTENTION BASED FEATURE EXTRACTION (IATEN)
Author:	K.VEENA, DR.V. SASIREKHA, S. DEVI
Abstract:	The brief duration of tweets makes it considerably more difficult to identify sarcasm, which is a difficult feature of sentiment analysis. The proposed model has effectively identified sarcastic attitudes in tweets while tested against a variety of Twitter datasets. It distinguishes itself from other cutting-edge models because of its capacity to accurately represent the sequential nature of language and concentrate on key words in the sentence. The LSTM model subsequently analyses the string of words, accounting for both the context and the word order, in order to precisely identify sarcasm. In comparison to conventional approaches that simply give every word in a phrase equal weight, the enhanced attention-based feature extraction used in this model is a significant advancement. In addition, the model can manage language's sequential characteristics due to the usage of LSTM, which makes it the best option for sentiment analysis employment. A significant advancement in the field of sentiment analysis is the proposed model for LSTM-based attention-based feature extraction and sarcastic sentiment analysis in tweets. To precisely identify sarcasm in text data, the model leverages the strength of LSTM with improved attention-based feature extraction. The framework is an excellent alternative for sentiment analysis tasks whereby sarcasm has to be recognized since it allows for the sequential flow of language and concentrate on important phrases in the sentence. To determine the sequential pattern of language while preserving the context and order of words in tweets, the model utilizes LSTM (long short-term memory), a sort of recurrent neural network. Furthermore, it uses an attention-based feature extraction technique that gives sentences key words greater weight. The algorithm can specifically detect linguistic nuance and discern sarcasm by concentrating on important words. In the proposed implementation, to correctly capture the sequential nature of language and concentrate on crucial words, it uses LSTM and better attention-based feature extraction, resulting in improved sarcasm recognition.
Keywords:	LSTM With Improved Attention-Based Feature Extraction, Twitter Datasets, Sarcasm, Sentiment Analysis
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	DIAGNOSING THE MEDICAL DATA USING ROUGH SET MIN- MAX CLASSIFIER
Author:	S. DEVI, Dr.V. SASIREKHA, K.VEENA
Abstract:	In today's medical disciplines, multiple massive quantities of data are being released, all of which constitute information on patients, diseases, and doctors. Disease The diagnosing process is one of the most important steps that requires a more costly examination. The illness outcome has been successfully predicted using a variety of different methodologies already in existence. However, it is far less capable of managing the huge and intricate medical dataset. A brand new Rough Set Min-Max Classifier (RMMC) is put to use in this approach that has been suggested in order to make illness forecasts. Based on the Euclidean distance measurement, the RMMC model describes the neighborhood connection between the two sets of instance data. This approach makes effective use of rough set theory, and the outcomes of the method's evaluation are examined using three distinct medical datasets. The experimental outcome of the RMMC technique that was presented is compared with the Neighborhood Rough Set Classifier (NRSC) algorithm, which stands for the Neighborhood Rough Set Classifier. The suggested technique achieves 99.42% accuracy in illness prediction, which is confirmed by the k-fold cross-validation, and has thus become a potential tool for diagnosing medical datasets. This is in comparison to the previous method, which only achieved 99.24% accuracy in disease prediction.
Keywords:	Rough Set, RMMC, NRSC, Euclidean Distance, Medical Diagnosis.
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	AI-DRIVEN PERSONALIZATION IN CUSTOMER RELATIONSHIP MANAGEMENT: CHALLENGES AND OPPORTUNITIES
Author:	Dr. N. VENKATESWARAN
Abstract:	Artificial Intelligence (AI) has emerged as a transformative technology with the potential to revolutionize various industries, including customer relationship management (CRM). This research study aims to explore the role of AI in enhancing CRM practices and improving customer experiences. This research paper discusses AI and CRM can analyze vast amounts of customer data, automate routine processes, and deliver personalized experiences on a large scale. By harnessing the power of AI, organizations can gain valuable insights into customer preferences, behavior patterns, and purchase history, allowing them to tailor their offerings and communications to individual customers. This paper explores the potential of AI in enhancing CRM strategies and improving customer experiences, ultimately leading to increased customer satisfaction and loyalty. The findings of this research will contribute to the growing body of knowledge on the application of AI in CRM and provide practical insights for organizations seeking to leverage AI technologies.
Keywords:	Customer Experience; Personalization; Customer Preferences; AI-Driven CRM; Accuracy
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	TWO STREAM SPATIAL-TEMPORAL FEATURE EXTRACTION AND CLASSIFICATION MODEL FOR ANOMALY EVENT DETECTION USING HYBRID DEEP LEARNING ARCHITECTURES
Author:	P. MANGAI, M. KALAISELVI GEETHA, G. KUMARAVELAN
Abstract:	Identifying events using surveillance videos is a major source that reduces crimes and illegal activities. Specifically, abnormal event detection gains more attention so that immediate responses can be provided. Video processing using conventional techniques identifies the events but fails to categorize them. Recently deep learning-based video processing applications provide excellent performances however the architecture considers either spatial or temporal features for event detection. To enhance the detection rate and classification accuracy in abnormal event detection from video keyframes, it is essential to consider both spatial and temporal features. Based on this, two-stream hybrid deep learning architectures like YOLOV4 with VGG16, Optical FlowNet with VGG16, and CNN-LSTM has been presented in this research work. The two-stream architecture handles the spatial features using hybrid YOLOV4, temporal features are handled using hybrid Optical FlowNet and finally, the features are concatenated and classified using CNN-LSTM based deep learning architecture. The proposed model attains maximum accuracy of 95.6% which indicates better performance compared to state of art of techniques.
Keywords:	Anomaly Event Detection, Keyframe Extraction, Spatial And Temporal Feature, Deep Learning, YOLOV4, Optical Flownet, VGG-16, Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM).
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	COMPARATIVE EVALUATION OF CARDIOVASCULAR DISEASE USING MLR AND RF ALGORITHM WITH SEMANTIC EQUIVALENCE
Author:	VINSTON RAJA R, DEEPAK KUMAR A, PRABU SANKAR N, CHIDAMBARATHANU K, THAMARAI I, KRISHNARAJ M, IRIN SHERLY S
Abstract:	Coronary artery disease is a highly intricate medical condition that affects a significant portion of the global population. It is even being referred to as a silent killer because it results in the death of a person with no obvious symptoms. The timely and accurate detection of heart disease is crucial in the healthcare industry, especially within the cardiology domain, as it enables effective treatment and management of the condition. Based on Machine learning techniques, an accurate and efficient model will be created to diagnosis heart disease. The machine learning models for classification will be developed using Multiple Linear Regression Algorithm and Random Forest Algorithm. Heart datasets were obtained from five countries: Cleveland, Hungary, Swiz, LongBeach, and Statlog, and datasets were analyzed using the Random Forest algorithm, KNN, Naive Bayes, SVM Algorithm and Multiple Linear Regression Algorithm to extract an intelligent pattern for forecasting the risk of heart disease. The accuracy of the MLR and RF models will be tested, and the best model will be deployed in health-care settings to diagnose cardiac disease.
Keywords:	Machine Learning, Random Forest (RF), Multiple Linear Regressions (MLR), Data Sampling
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	STEGANALYSIS OF DATA HIDDEN IN RGB IMAGES USING RESIDUAL NEURAL NETWORK
Author:	ALANOUD MAHMOUD ALMASAR, OSAMA MAHMOUD OUDA
Abstract:	Digital forensic analysis aims to apply scientific and statistical techniques to identify, gather, preserve and present relevant digital evidence which will allow to confirm or reject a hypothesis against possible criminal activity. Current methods of forensic digital analysis are effective for visual analysis of physical evidence, but they do not allow for the automatic execution of large amounts of data, for correlation studies of the files obtained, for the validation of the metadata and for the identification of abnormalities in text, graphic or audiovisuals files. For this reason, artificial intelligence techniques were proposed for processing data, for identifying patterns and trends that make it possible to perceive aspects that cannot be visually perceived. In this paper, we propose a network analysis approach to steganalysis of RGB images in frequency domain based on Residual Neural Network-50 (ResNet-50). We used the Discrete Cosine Transform (DCT) features based on standard convolutional operations in ResNet-50 generated by the first convolutional layer in the network. The DCT Feature based ResNet-50 model extensively reduces the quantity of parameters and multiplications while keeping comparable accuracy results to normal ResNet in ImageNet-1K.
Keywords:	Steganalysis, Deep Learning, Convolutional Neural Networks (CNN), Forensics.
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	WEBSHELL DETECTION BASED ON BYTECODE FEATURE WITH CONVOLUTIONAL NEURAL NETWORK
Author:	DIAN ANGGRAINI, ABBA SUGANDA GIRSANG
Abstract:	Web shell is a malicious program used to remotely access web servers during cyberattacks. Malicious web shells closely resemble benign web shells, making them difficult to distinguish. The challenge in detecting pre-existing web shells is that this type of malware is hard to detect using an intrusion detection system (IDS) or antivirus techniques. This is because web shells are usually hidden within web applications, making them challenging to differentiate from regular web application source code. Therefore, traditional detection models that analyze the dynamic features of web shell script execution are more effective in detecting existing malware attacks. In this study, A method of web shell detection based on dynamic bytecode features using a convolutional neural network (CNN) has been proposed in this research. Word2vec is employed to obtain vectorized features from the bytecode or opcode. Experimental results using a training dataset of 2577 samples and a validation dataset of 645 samples yield the best model with an accuracy of 99.86% at epoch 100. The experiments demonstrate that this model effectively detects web shells, with a significant increase in accuracy levels.
Keywords:	Web Shell, Machine Learning, CNN, Cyber Security, Opcode.
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	MACHINE LEARNING MODEL TO IMPROVE CLASSIFICATION PERFORMANCE IN THE PROCESS OF DETECTING PHISHING URLS IN QR CODES
Author:	BETTY HERLINA, HARYONO SOEPARNO
Abstract:	This paper discusses the model of detection of phishing URLs in QR codes using Machine Learning techniques. Phishing URLs are URLs that resemble real websites created by cybercriminals to obtain user information for profit. The dataset is collected from various sources, with some important features based on address bar, domain and HTML. The six Machine Learning algorithms including Decision Trees, Random Forests, SVM, Multilayer Perceptron, Autoencoder Neural Network and XGBoost. The experimental results show that the XGBoostt algorithm provides the highest detection accuracy with 92% accuracy.
Keywords:	Urls Phishing, Qr Code, Python, Machine Learning, Security Cyber.
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	OPTIMIZING PERFORMANCE OF COVERAGE DELAY IN WIRELESS MOBILE NETWORKS USING POWERED CLUSTER BASED ROUTING
Author:	R. SARAVANKUMAR, T. TAMILSELVI, K. REVATHI, S. J. VIVEKANANDAN, W. GRACY THERESA
Abstract:	Delay/Disruption Tolerant Networking (DTN) permits the effective usage of numerous pathways and providers. The DTNs are different because they are not linked to each other. This means they don't have direct paths from one end to the other. Popular ad hoc routing protocols like Ad hoc on-demand distance vector (AODV) and dynamic source routing (DSR) are unable to construct routes in these difficult settings. Clustering is isolating the similar mobile nodes into numerous strewn groups, named as clusters. The transmission of data is done between various clusters in a cluster-head-to-cluster-head (CH-to-CH) fashion. In this research paper a novel clustering based approach is investigated for DTN, which is unique and non-trivial as the network links exist only temporarily, which makes it impossible to achieve end-to-end connection for data delivery. Therefore routing is based on nodal contact probability. Here, an exponentially weighted moving average (EWMA) method is applied, which is mainly for the online updating of nodal contact probability. Though the result gives an improved packet delivery ratio through which a high throughput and minimum energy consumption, this increases the burden on every CH. This study discusses an ideal clustering technique that uses two mechanisms: optimal cluster planning with routing awareness and optimum random relay with clustering awareness – to balance the strength of each cluster in a stochastic condition. As an outcome, optimal power consumption is attained with increased throughput and packet delivery ratio.
Keywords:	Delay Tolerant Network, Clustering, Coverage Constraints, Delay, Performance, Routing, Wireless Sensor Networks
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	IMPROVING DETECTION SQL INJECTION ATTACKS USING REFERRER HEADER AND URI WEBLOG
Author:	ELFRIDA B.A SIAHAAN, ABBA SUGANDA GIRSANG
Abstract:	SQL Injection Detection provides the ability to monitor SQL Injection attacks on websites. Currently, researchers are using Deep Learning to detect SQL Injection. However, this detection has limitations, such as high False Positives (FP), False Negatives (FN), and low Accuracy due to the detection of SQL Injection using only URI data. At the same time, attacks do not only occur through URIs but also Referrers. Therefore, this study aims to use the combination of URI and Referrer to detect attacks and discuss the increase in model performance due to adding a Referrer. This study's first step was preprocessing the dataset and then vectorizing it using Word2Vec. The Word2Vec and CNN method was proposed using the combination of URI and Referrer, then compared to Word2Vec CNN using URI. The experimental results show that the proposed method performs better than other methods and gets accuracy over 99% of the payloads with a low error rate.
Keywords:	CNN, Referrer, SQL Injection, Web Log, Word2Vec
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	DETECTION OF ANOMALIES IN BLOCKCHAIN USING FEDERATED LEARNING IN IOT DEVICES
Author:	SWAPNA SIDDAMSETTI, DR. MUKTEVI SRIVENKATESH
Abstract:	Botnet attacks now pose a substantial cyber security threat to the Internet of Things (IoT). Botnet classification systems that depend on these methodologies, like ordinary machine learning and deep learning, cannot be scaled. Based on federated learning (FL), researchers have created a classification approach for botnet attacks. This paper aims to solve the challenges of safeguarding user privacy while attaining acceptable classification performance. However, even a single, underperforming local model is included in each round's global model. In that case, the traditional FedAvg may produce an underperforming global model. FedAvg assigns equal weight to all local models when determining the average. This study develops dynamic weighted updating federated averaging (DWU-FedAvg) to overcome this problem. A system that dynamically modifies local model weights depending on client performance is needed to accomplish this goal. When the DWU-FedAvg is tested, it is compared to two well-known benchmark datasets, BotIoT and N-BaIoT. Both of these datasets are used in botnet attack classification studies. According to the results, our proposed model is scalable and capable of protecting user privacy while beating the classical FedAvg in terms of accuracy, with 98.4% for 15 rounds and 98.9% for 20 rounds for Botnet attack classification.
Keywords:	Privacy preservation, FedAvg, DWU-FesAvg, Federated Learning, Blockchain
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	ALGORITHMICALLY GENERATED MALICIOUS DOMAIN DETECTION USING N-GRAMS EMBEDDING AND ATTENTION-BASED BIDIRECTIONAL GATED RECURRENT UNIT
Author:	ALFONSUS SUCAHYO HARIAJI, ABBA SUGANDA GIRSANG
Abstract:	Botnets are one of the recent main cyber security threats. In order to avoid detection, botnets use Domain Generation Algorithm (DGA) to generate malicious domain names and maintain communication between infected bots and command and control server (C&C). Botnet malwares use various algorithm to generate domain names such as arithmetic, hashing, and wordlist/dictionary techniques. Recent traditional machine learning and deep learnin based detection methods need handcrafted domain name features which require more effort and advanced expertise and knowledge. This study aims to detect and classify DGA malicious domain without manually define and handcraft domain name features by only using the domain name. N-grams method was used to create sequences of domain names and then vectorize the sequences using word embedding technique to create n-grams embedding model. After vectorization, Bidirectional Gated Recurrent Unit (BiGRU) was used for domain name classification and attention mechanism was used to improve classification performance by applying attention weight. The experiment results demonstrate the N-Grams Embedding and Attention-based BiGRU model proposed in this paper can detect and classify various type of DGA domains generated by arithmetic, hashing, and wordlist algorithm more effective compared to older algorithm such as CNN and LSTM for both DGA malicious domain detection and classification task. The use of attention mechanism can also improve the accuracy and performance of the DGA malicious domain detection model compared to models that do not use attention mechanism.
Keywords:	Attention Mechanism, Domain Generating Algorithm, Gated Recurrent Unit, Malicious Domain, N-grams Embedding,
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text

Title:	CLASSIFICATION OF TRAINING ENGINEERING RESULTS USING DIGITAL TOOLS: CASE OF PATIENTS TREATED BY DIALYSIS METHOD
Author:	JAOUAD CHOUIKH, NADIA OUZENNOU, AIT EL HAJ SALOUA, NEZHA NACER1 AND SAMIA RKHA
Abstract:	Data classification consists of organizing them into categories according to agreed criteria. Carefully planned classification enables more effective use of critical data and its protection across the enterprise; it also participates in risk management and legal knowledge and compliance processes. There is no universal approach to data classification. However, the classification process can be broken down into four key steps, which you can customize to meet your specific business needs when developing your data protection strategy. In this article we have tried to classify the results of training engineering by taking into consideration professional training carried out with health care personnel on dialyse. The subject focuses on the classification of training engineering results using numerical tools, focusing on patients treated by the dialysis method. The aim is to categorize outcomes related to training using digital technologies, based on data from patients undergoing dialysis treatment. This approach aims to identify patterns and trends in outcomes for a better understanding of patient needs and progress. The use of digital tools makes it possible to efficiently collect and analyze medical, laboratory and quality of life data from patients, in order to classify the results according to their health status, their response to treatment and their evolution over time. This methodology offers advantages for personalizing care, anticipating patient needs and improving the overall management of patients on dialysis.
Keywords:	Classification, Training Engineering, Digital Tools, Perineal Rehabilitation
Source:	Journal of Theoretical and Applied Information Technology 30^th September 2023 -- Vol. 101. No. 18-- 2023
Full Text