		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

February 2023 | Vol. 101 No.3

Title:	CHILDRENS HAPPINESS, ENJOYMENT, PERCEIVED MOTIVATION AND ACHIEVEMENT TOWARDS SCIENCE AUGMENTED REALITY PICTURE BOOK
Author:	NUR AADILA AHMAD RAZI, NURULLIZAM JAMIAT
Abstract:	This study aims to explore the application of an augmented reality (AR) picture book in science subjects among children. A mixed-method research design was adopted where qualitative and quantitative data were collected and corroborated to address the research questions. There were 60 children aged-six years old with 33 boys and 27 girls that were involved in this study. Data were collected using surveys, interviews, and pre and post-test questions. The findings showed that children were happy to learn science using the AR picture book due to the vibrant colours and animation portrayed. Children's level of happiness was found to have a significant and positive relationship on their achievement. These results provide a promising perspective of AR technology for children’s learning and contribute to the scarce study on AR among children. Future research on the use of AR technology for other subjects or younger children could be further explored to provide insight into children's learning through AR as an early childhood intervention.
Keywords:	Augmented Reality; Children, Happiness; Enjoyment; Perceived Motivation; Achievement; Science Augmented Reality Picture Book
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	A NOVEL HOMOMORPHIC AND MATRIX OPERATION FOR RANDOMIZATION ENCRYPTION SCHEMES FOR PRIVACY IN CLOUD COMPUTING ARCHITECTURE
Author:	R. HARI KISHORE, A. CHANDRA SEKHAR, PRAMODA PATRO, PRAGATHI CHAGANTI
Abstract:	Traditional outsourced methods are being replaced by the emerging cloud computing architecture, which offers adaptable services to clients in many regions through the Internet. Due to this, it becomes necessary for data categorization to be carried out by probably dangerous cloud servers. In this situation, clients can use a classification created by the server to categorize their unique cloud-based resource test results. The current study effort created a training approach that protects confidentiality using the Matrix Operation for Randomization and Encryption (MORE) scheme, allowing neural network model calculations. The neural unit of the model may retain certain critical data as a result of over-fitting during the deep learning training phase. It is possible for data privacy to be compromised whenever the intruder creates the appropriate assault scenario. In this research article, a Differential Privacy Subspace Approximation with Adjusted Bias (DPSaab) is proposed to train the Feed forward-designed Alexnet convolutional neural network (FF-ACNN). Initially, the data normalization is done by Min-Max normalization then the privacy-preserving training method based on A FF-Alexnet CNN model's calculations may be done immediately on floating point data. This is possible due the encryption technique MORE. In addition, the DPSaab method is used to maintain differentiated anonymity in FF - CNN's Alexnet. The efficacy of the suggested approach is tested on the MINST identification database using the DPSaab method, which meets the concept of differentiated secrecy.
Keywords:	Classification Problem, Homomorphic Encryption, MORE, MNIST Digit Recognition Problem, DPSaab, FF-ACNN.
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	PERFORMANCE ANALYSIS OF PLASTIC INJECTION MOLDING USING PARTICLE SWARM BASED MODIFIED SEQUENTIAL QUADRATIC PROGRAMMING - ALGORITHM MULTI OBJECTIVE OPTIMIZATION MODEL
Author:	EKTA S MEHTA , Dr S.N. PADHI
Abstract:	Plastic Injection Molding (PIM) is one among the trending technology which is used in maximum of the plastics based industrial applications. The functionalities of PIM are material insertion, melting, injection, molding closing, mold ejection and cooling. Several variables are involved in the process of molding such as melt and mold temperature, melting and injection time, packing pressure and cooling time. In order to achieve high quality, effective performance in terms of maximizing the weld line temperature and reduction of clamping force. To attain this it is very essential to optimize the variable selection during PIM process. For that purpose, in this research Multi Objective Optimization (MOO) is proposed in plastic injection molding which is the combination of Particle Swarm Optimization (PSO) and a Modified Sequential Quadratic Programming Algorithm (MSQPA) as well as finally using the Ant–Lion Optimization algorithm our proposed PSO-MSQPA is compared. The simulation of this process is performed using the software such as CATIA V5 and MATLAB. Through the numerical and simulation evaluation of the proposed PSO-MSQPA approach, the variables are effectively optimized during the process of injection molding that leads to improve the convergence and reduces the bit error rate and that leads to improve the quality by reducing the clamping force of the PIM process.
Keywords:	Plastic Injection Molding (PIM), Multi Objective Optimization (MOO), Particle Swarm Optimization (PSO), Sequential Modified Sequential Quadratic Programming Algorithm (MSQPA), Ant-Lion Optimization Algorithm
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	RISK ASSESSMENT RELATED TO PRIVACY INFORMATION ON ELECTRONIC MONEY SERVER-BASED USING ISO 27001 ISO 27005, ISO 27701
Author:	SYIFAURACHMAN, ANTONI WIBOWO
Abstract:	Electronic money server-based issuers are faced with the reality of the emergence of threats from information technology risks and insufficient knowledge of the impact of the risk of user data leakage on electronic systems used. However, on the other hand, the operator does not have an integrated method to identify and assess the risk to information technology security and user data privacy. This study focuses on the integration risk assessment of information security and user data privacy in electronic money server-based mobile applications using ISO 27001:2013, ISO 27005:2018, and ISO 27701:2019. The data was obtained from one of the providers in Indonesia through leadership interviews and observations of mobile applications with the scope of business processes for user registration, top up of balances, and acting as a Personally Identifiable Information Controller. The evaluation uses the KAMI Index version 4.1 for understanding the condition of the organization and the final results after the implementation of the risk assessment based on the three implemented standards. The results of the study explain that company XYZ has an electronic system category with a value of 26 (High), and the value of the application of information security and privacy increases from level I (initial condition) to II (basic framework). This study provides a comparative analysis and method of information technology risk assessment related to privacy information, so that the use of the three proposed standards can be a reference for any organization to expand the control of information technology security controls on user data privacy and compliance with regulations.
Keywords:	Risk Assessment, Information Security, Privacy Information, Electronic Money
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	ADDICTION ANALYSIS AND NEUROTICISM OF USE OF INSTAGRAM SOCIAL MEDIA ON PRODUCTIVE AGE IN JAVA ISLAND
Author:	YUDHISTIRA AUDRI PRANATA, RIYANTO JAYADI
Abstract:	Social media has become an inseparable part of people's lives. People share their daily activities, experiences and opinions on social media. Social media has characteristics that require an extraordinary approach. Instagram is one of the most popular social media in Indonesia. This study aims to determine the correlation between Instagram usage and Neuroticism through excessive use of Instagram and can provide important information related to Instagram usage addiction to Instagram users. Neuroticism is described through a person's personality, such as the emergence of individualism, anxiety, fear, worry, irritability, FOMO, insecurity about other people's achievements, feeling inferior, becoming mentally unhealthy and forgetting responsibility. Many people complain that it is difficult to finish when using Instagram, and even their way to end it is by doing useless things. The method used is quantitative by distributing questionnaires to 400 productive-age respondents in Java Island. After that, the Structural Equation Model (SEM) was calculated using Smart PLS. The research model used in this study is a modification of several models, such as Big Five Personality Traits, DeLone and McLean, the Technology Acceptance Model, the Unified Theory of Acceptance and Use of Technology 2, and Lifestyle. The findings are that Information Quality, System Quality, Performance Expectancy, and Habit affect Instagram usage, while Perceived Ease of Use does not affect Instagram usage. In addition, it was also found that using Instagram affects a person's Neuroticism.
Keywords:	Big Five Personality Traits, DeLone and McLean, Instagram, TAM, UTAUT2.
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	MANGROVE FRUIT RIPENESS CLASSIFICATION USING DEEP CONVOLUTIONAL NEURAL NETWORK
Author:	SHARFINA FAZA, ROMI FADILLAH RAHMAT, MERYATUL HUSNA1, RINA ANUGRAHWATY, AJULIO PADLY SEMBIRING, RHAMA PERMADI AHMAD, ONRIZAL
Abstract:	Mangrove is a community of plants that live between the sea and land which is affected by tides. Indonesia has the largest mangrove forest in the world and also has the largest biodiversity and most varied structure. In general, mangrove plants have several benefits, such as preventing coastal erosion, preventing seepage of seawater to land which can cause groundwater turns into turbid condition, as a place to live and a source of food for several species of animals. Mangrove plants consist of several parts, from the stem of the tree leaves, flowers and fruit. To get the optimal mangrove plants, it is necessary to have fruit which is optimal based on the ripeness level. In general, by assessing the ripeness of mangroves is only by looking manually with the eyes, so that the accuracy of mangrove fruit ripeness valuations isn't high because of they only see with eyes. Many mangrove farmers think that mangrove want to plant in case of rehabilitation, already have a good level of ripeness, but after replanted, the result isn't appropriate. To overcome this case, this study utilizes digital image processing using the Deep Convolutional Neural Network method to help the communities and farmers in recognizing the ripeness level of mangrove fruit. The image processing techniques used in this study are Grayscaling, Adaptive Threshold, Sharpening, and Smoothing. After tested in this study, it was concluded that the method applied can determine the ripeness of mangroves well and the accuracy obtained is equal to 99.1%.
Keywords:	Mangrove, Grayscaling, Sharpening, Adaptive Threshold, Deep Convolutional Neural Network
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	THE APPROACH USING CUMULATIVE VOTING AND SPANNING TREE TECHNIQUE IN IMPLEMENTING FUNCTIONAL REQUIREMENT PRIORITIZATION: A CASE STUDY OF STUDENT’S FINANCIAL SYSTEM DEVELOPMENT
Author:	NORHAWANI AHMAD TERIDI, ZAHRUL AZWAN ABSL KAMARUL ADZHAR, NASRUDIN MD RAHIM3, JAMILAH KAMIS, TAUFIK RIDZUAN5, ZURAIDY ADNAN, MUHAMMAD FAIRUZ ABDUL RAUF
Abstract:	Requirement Prioritization (RP) occurs while gathering the user requirement of system development. This is to ensure the process for eliciting requirements meets the business flow as well as project management such as timeline and cost. The techniques used in RP are a combination of ratio and ordinal scales that are cumulative voting and minimal spanning tree techniques. The significance of RP is to improve project development time which may require many months or several years, therefore it is essential to determine the requirements that should be implemented at the beginning. From the result, the RP combination technique can be applied in identifying the priority of functional requirements for the case of developing a student's financial system.
Keywords:	Requirement Prioritization, Functional Requirement, Cumulative Voting, Spanning Tree, Student Financial System.
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	ALGORITHM FORWARD CHAINING AND BACKWARD CHAINING FOR DIAGNOSING DIABETES MELLITUS
Author:	HANDRIZAL, HAYATUNNUFUS, MUHAZIR FANDI
Abstract:	Diabetes Mellitus is a chronic disease or disorder with various causes characterized by high blood sugar levels accompanied by impaired carbohydrate, lipid, and protein metabolism due to a lack of insulin function. There are three types of diabetes mellitus, namely diabetes mellitus type I (the human body fails to produce insulin), type II (cells fail to use insulin), and gestational (high blood sugar levels during pregnancy). Diabetes Mellitus is a disease that is often underestimated by some people due to the lack of knowledge about diabetes mellitus them knowing diabetes mellitus is a disease that must be considered because it can cause death slowly, especially in the elderly and pregnant women. It is necessary to build an expert system that can diagnose diabetes mellitus, the system will be built using forward chaining and backward chaining algorithms, and can function to diagnose and provide good treatment for diabetes mellitus. An expert system is a specialized piece of software or computer program, that stores expert knowledge about a particular problem domain, often in the form of IF-THEN rules that are capable of solving problems at a level equal to or greater than that of human experts. With this expert system, even ordinary people can solve quite complex problems that can only be solved with the help of experts. Then for doctors, this expert system is also very helpful in their activities as very experienced assistants. The data used in this test are 50 data for diagnosis and 50 data for treatment, so the total data is as many as 100. The results of the comparison test all data match the results of expert test data and system results. Then the level of accuracy of this system is 100%. The results of this study succeeded in increasing the accuracy of previous studies, namely from 90% to 100%.
Keywords:	Diabetes mellitus, Expert System, Forward Chaining, Backward Chaining
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	EFFICIENCY OF SURVEILLANCE OF TCP PACKET IN IOT IN REDUCING THE RISK OF RANSOMWARE ATTACKS
Author:	RANA ABDUL SAMI KHAN , PROF DR. MOHD.NORDIN ABDUL RAHMAN
Abstract:	Introduction: In recent times, up-to-date IoT systems have implemented the open standards of Transmission Control Protocol (TCP) practices in order to accommodate the heterogeneity of applications and devices from various sellers. The study aims to understand the efficiency of surveillance of TCP packets in IoT in reducing the risk of ransomware attacks. More specifically, the study is focused on determining the efficiency of surveillance of TCP packet used in IOT. Meanwhile, investigating the effectiveness in reducing the risk of ransomware attacks through surveillance TCP in IOT has also been one of the present investigation's objectives. Findings: The findings of the study suggests that IoTSDN-RAN is proven as effective surveillance technique used in reducing the risk of ransomware attacks. The technique has been considered as efficient when is compared with the IoT SVM and IoT ML techniques. In terms of detection, precision and accuracy, the technique was proven as efficient in comparison to the other most frequently utilised techniques. The challenge in SVM is to find the support vectors, to classify unknown traffic to the attacks. Recommendations: The users need to understand that IoT tends to be a complex and thus, they need to have a better understanding with regards to the TCP and ransomware attacks and avoid them through the TCP implications.The implications of TCP by the IT users is important to ensure that there is effectiveness and efficiency addressed among the entire system and there pertains no risk of ransomware attacks.
Keywords:	TCP Packet, Ransomware Attacks, Iotsdn-RAN, Iot SVM, Iot ML.
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	A DEEP LEARNING BASED TECHNIQUE FOR THE CLASSIFICATION OF MALWARE IMAGES
Author:	MD. HARIS UDDIN SHARIF, NASMIN JIWANI, KETAN GUPTA, MEHMOOD ALI MOHAMMED, DR.MERAJ FARHEEN ANSARI
Abstract:	Because of the fast expansion of the internet and technology, a slew of developing malware and attack techniques has evolved. As a result, researchers concentrated their efforts on machine learning and deep learning techniques to detect malware. Many organizations have been developing new algorithms and products to secure people from these scams. On the other hand, Malware kinds have been expanding substantially in recent years. The anti-virus companies have been discovering millions of new malware variants every year. Therefore, new intelligent malware detection methods must be solved as soon as possible to halt this rise. Malware is becoming more prevalent, more diverse, and more sophisticated. Deep learning in malware detection through images has recently been demonstrated to be highly effective. We also employed an Image-based Malware dataset [Malimg] and used the different deep learning algorithms, CNN, Caps-Net, VGG16, ResNet, and InceptionV3, for malware detection. The dataset images were transported through the pre-processing pipeline and into the deep learning pipeline, where they were used to train deep learning models in the right way. As part of the model training process, all images were resized to be the same size and proportions. A factor of 1/255 was then applied to the images, resulting in a conversion from RGB value to grayscale, which restored the original RGB values to their correct positions. Later, the dataset was segmented into two groups, train, and test. The VGG16, ResNet50, and InceptionV3 models detected the malware images. A combination of the Adam optimizer and the cross-entropy loss function was used to train all of the models. The models were trained for 50 epochs using early stopping criteria. Finally, the model composition method was used to classify malware images where the previously trained models were combined. The custom CNN model, the VGG16, ResNet50, and InceptionV3 models were combined to predict a single outcome for the experimental condition. The proposed technique provided very promising results.
Keywords:	Malware Prediction, VGG16, ResNet50, Caps-Net, Image-Based Malware Prediction, Cyber Analysis, Deep Learning, Cyber Security
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	BLOOD CELLS CLASSIFICATION USING DEEP LEARNING WITH CUSTOMIZED DATA AUGMENTATION AND EK-MEANS SEGMENTATION
Author:	ASHRAF HENI, IMEN JDEY, HELA LTIFI
Abstract:	White blood cells, also known as leukocytes, play an important role in the human body. By increasing immunity for fighting infectious diseases. The classification of white blood cells plays an important role in detecting disease in an individual. Classification can also help in the identification of diseases such as infections, allergies, anaemia, leukaemia, cancer, Acquired Immune Deficiency Syndrome (AIDS), etc., which are due to abnormalities of the immune system. Currently, there is a great deal of research being done in this area. In some cases, the classification of white cells is a medical emergency that requires rapid diagnosis. given the complexity of classifying them into subtypes, researchers have presented new techniques to help healthcare workers better facilitate this task. scientific reviews have shown the performance of classification systems using deep learning. We will use a deep convolutional learning technique (CNN) that can classify WBC images into its subtypes namely neutrophils, eosinophils, lymphocytes, and monocytes. The proposed approach will rely on data augmentation techniques for classification and the introduction of the EK-means algorithm for image segmentation which is a fusion between k-means and fuzzy c-means. we introduce a new technique for data augmentation called checkerboard image. this technique will be introduced in the context of further model learning by simulating the pixelization of images or hole images. By applying the emerging deep learning-based image classification technique "convolutional neural network" using Ek-means for segmentation and data augmentation for classification to a large database, we were able to classify WBC, using VGG19 model, into its subtypes with a validation accuracy of 96,24%. We also compare our proposed model with the reference models.
Keywords:	White Blood Cells, EK-Means, VGG19, CNN, Data Augmentation
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	COMPARISON OF THE ACCURACY OF THE LEXICON-BASED AND NAIVE BAYES CLASSIFIER METHODS TO PUBLIC OPINIONS ABOUT REMOVING MASKS ON SOCIAL MEDIA TWITTER
Author:	ANGGA ADITYA PERMANA, WAHYU ALDHI NOVIYANTO
Abstract:	At the beginning of 2020 the world was shocked by the COVID-19 pandemic which paralyzed all aspects of activity for some time. However, over time and with the discovery of a vaccine, the cases caused by COVID-19 began to subside. In 2022, the Indonesian government make a policy that people are allowed to take off their masks when active but are encouraged to maintain health protocols. However, the approach reaped the pros and cons of the Indonesian people. One challenge is to build technology to detect and summarize an overall those pros and cons. So that, we look at Twitter and build models for classifying ‘tweets’ into positive, negative and neutral sentiment using top two approaches for sentiment analysis, the lexicon-based method and the naive Bayes classifier. This study aimed to analyze public opinion about removing masks through Twitter by comparing the lexicon-based method and the naive Bayes classifier method to find out how the community responded to taking off masks. A total of 639 tweets with the keyword "Lepas Masker" was analyzed include data crawling, text preprocessing, feature extractions and the classification process. The comparison of the results obtained shows the accuracy of 82% for the lexicon-based method and 70% for the naive Bayes classifier method. To the results, the accuracy value of the lexicon-based method is higher than the naive Bayes classifier method.
Keywords:	Analysis Sentiment, Covid-19, Removing Mask, Naïve Bayes, Lexicon.
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	EVOLUTIONARY AUTO ENCODER TECHNIQUE FOR DETECTION OF OVERLAPPING COMMUNITY IN SOCIAL NETWORK WITH INTERPRETABILITY USING AEOCDSN ALGORITHM
Author:	PRANAVATI BAJRANG JADHAV, DR VIJAYA BABU BURRA
Abstract:	Social networks have increased and are gaining popularity because of the Internet's rapid technological advancement. Researchers' interest in a study on numerous social networks is growing. Community detection is crucial for social networks regarding issues with online security, such as user behavior analysis and anomalous community detection. The unique feature of a social network community is the multi-membership of a node, which leads to overlapping communities. However, solving the overlapping community’s detection issue costs much computing. In social networks, communities develop as a result of nodes' self-interest. The social network's nodes play the role of self-interested people who want to get the most from interactions as communities. An Auto Encoder for Overlapping Community Detection in Social Network (AEOCDSN) is suggested in this paper to detect the overlapping community in the social network. This detection helps in recommending and understanding user interest in different matter. Low-dimensional vertices models are learned using Label Propagation Algorithm (LPA) algorithms and community input in the suggested model. The modularization architecture makes the system more adaptable with a lesser loss function. Overlapping community detection techniques are used to gather community data in the social network. In addition to effectively integrating and preserving data from regional and global viewpoints, LPA and Infomap approaches can improve the models' robustness.
Keywords:	Sparse Auto Encoder, Overlapping Community, Social Network, LPA, Infomap
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	INVESTIGATION OF INTRUSION DETECTION SYSTEM USING RANDOM FOREST, CART AND PROPOSED SECURE RANDOM FOREST ALGORITHMS (SRFA)
Author:	P.KANAGAVALLI, DR.V.KARTHIKEYANI
Abstract:	Wireless Sensor Network (WSN) are composed of low cost sensor nodes and commonly deploy in open and unprotected area, which make security the principal mission in this form of network, due to their traits WSN is susceptible to various types of attacks and intrusions, the place it require security mechanisms to protect towards these attacks. Intrusion detection machine (IDS) is one of the major and efficient protecting strategies against intrusion and attacks in WSN. In this paper, a novel feature extraction algorithm, specifically Correlation Based Feature Extraction (CFS) algorithm is proposed with an aim to limit the training time and to decorate the lifetime of the system. The trust level node is estimated by using utilizing the behavior analysis and residual energy level of nodes. Thus, we have proposed a new Trust Algorithm (TA) to compute the trust level of nodes in the network. Finally, SRFA primarily based classifier is used to classify the nodes into a trustworthy, untrustworthy or malicious node based totally upon the measured trust stage of the nodes. The results absolutely confirmed that the proposed intrusion detection system extensively reduces the false positive rate, thereby proving that the proposed approach is capable of identifying anomalies in network better than different current system. As cyber threats develop in sophistication, network defenders need to use every device in the defensive arsenal to defend their networks. Data mining techniques, such as decision tree analysis, provide a semi-automated approach to detect adversary threats. In this paper, a novel feature extraction algorithm, specifically Correlation Based Feature Extraction (CFS) algorithm is proposed with an intention to limit the training time and to beautify the lifetime of the system. The trust level node is estimated via using the behavior analysis and residual energy level of nodes. Thus, we have proposed a new Trust Algorithm (TA) to compute the trust level of nodes in the network. Finally, SRFA based classifier is used to classify the nodes into a trustworthy, untrustworthy or malicious node based totally upon the measured believe degree of the nodes. This proposed method is compared with SVM and C4.5 and evaluates the overall performance the usage of KDD99 dataset. Simulation results prove that the proposed SRFA can successfully mitigate malicious node and gives higher effects when in contrast to SVM and C4.5. In previous years, a dramatic enhancement issues in the number of attacks, intrusion detection field and it becomes the mainstream of data assurance. Sensors nodes are used in WSN with the onboard processors that manages and monitors the environment in the a particular area. They are connected to the Base Station which acts as a processing unit in the WSN System. In WSN, data mining is the process of extracting model and pattern that are application oriented with possible accuracy from rapid flow of data. Because of their special characteristics, and limitations of the WSNs, the traditional data mining approaches are not directly applicable to WSNs. A widespread analysis of different pre-existing data mining techniques adopted for WSNs are examined with different classification, evaluation approaches in this paper. Finally, a few research challenges to adopt data mining methods in WSNs are also pointed out. A general concept of how traditional data mining techniques are improved to attain better.
Keywords:	Decision Tree, Intrusion Detection System, SVM, C4.5, CART, Secure Random Forest Algorithms
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	AN EFFICIENT ELECTRONIC HEALTH RECORD (EHR) BASED INTEGRITY AND MCP-ABE MODEL FOR DISTRIBUTED BLOCK CHAIN FRAMEWORKS
Author:	KEESARA SRAVANTHI, P CHANDRA SEKHAR
Abstract:	Most of the conventional cloud based security frameworks are applicable to provide data security on homogeneous electronic health records(EHRs) due to data size and computational time. In the unstructured EHRs, traditional models use limited data size with partial data security in cloud computing environment. In this work, a hybrid security framework is proposed in order to provide strong security to the block chain heterogeneous databases. In this work, homogeneous 2D EHRs and heterogeneous 3D EHRs are used to compare the proposed approach to the traditional approaches. Experimental results present the proposed heterogeneous data security framework has better efficiency than the conventional models in terms of computational overhead and average runtime(ms) computations.
Keywords:	EHR Data, Cloud 2D Data, Medical 3D Data, Attribute Based Encryption.
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	AN ADAPTIVE APPROACH TOWARDS PREDICTION AND DIAGNOSIS OF LUNG CANCER USING HEURISTIC GREY WOLF OPTIMIZATION APPROACH AND RANDOM FOREST CLASSIFIER–LUCAGO
Author:	A.ANUPRIYA, ARUNKUMAR THANGAVELU
Abstract:	Lung Cancer is considered to be a most disastrous health disease, where chances of recover is minimal as per World Health Organization. This research work supports on early prediction of lung cancer, which do help on chances of patients under risk to be diagnosed. Proposed work LUCAGO, focuses on understanding the chances of cancer attack and determining lifetime of patient whose prediction of lung cancer, which demands accuracy on prediction quality.LUCAGO frame work follows adaptive approach towards prediction of lung cancer and understanding the stochastic growth of cancer tumour cells using GWO for feature selection and random forest as classification approaches. LUCAGO approach adopts the knowledge of stochastic cancer symptoms as predictable patterns to suggest on prediction among detectable patient’s using unstrained dataset.The performance evaluation of LUCAGO method demonstrates positive results, where GWO can be adopted effectively adopted by clinical research community oncologists to support on early identification of lung cancer. Work being verified over analytical metrics such as accuracy, error analysis and time taken for prediction shows that LUCAGO is far better in performance compared to GA, ACO approaches.
Keywords:	Lung Cancer, Early Prediction, GWO, Random Forest.
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	ENHANCING STANDARD ENCRYPTION ALGORITHMS USING MULTILAYERS ENCRYPTION TECHNIQUE
Author:	SHAIMA ALFADHLI, KHOLOUD SAK, MOHAMMED ALWAKEEL
Abstract:	There are several standard encryption algorithms used to encrypt text to protect it from unauthorized access, however due to advance in technology it has become easier for attackers to decrypt the encrypted text and gain an unauthorized access to the original text. Therefore, it has become a necessity to develop new schemes that increase the level of security of the standard encryption algorithms. In this research, we will introduce a new multilayers encryption/decryption technique that can be used to enhance any standard encryption algorithm. In particular, the proposed technique consists of four layers, in the first layer, the original text is anonymously randomized using a private randomization key, then in the second layer, a standard encryption algorithm is used to encrypt the randomized text resulting from the first layer, in the third layer, the features of a private image (image that is known only by the transmitter and the receiver) are extracted using artificial intelligent technique, and then these features are used as a private key to encrypt the text resulting from the previous layer, finally, in the fourth layer, the encrypted text resulting from the third layer will be hidden in a cover image using standard steganography technique and transmitted to the receiver. At the receiver side, the same steps will be executed in reverse order to get the original text. The proposed technique is expected to increase the security level of the standard encryptions algorithm, and makes it more complex for the attacker to decrypt the text.
Keywords:	Encryption Algorithm, Multilayers Encryption, Anonymous Randomization, Text Randomization, Image Steganography
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	AN EFFICIENT GROUP KEY AGREEMENT PROTOCOL IN TELEMEDICINE
Author:	SRAVANI JAYANTI, K CHITTIBABU, PRAGATHI CHAGANTI, CHANDRA SEKHAR AKKAPEDDI
Abstract:	Telecommunication is an important trait in telemedicine. In an e-Health Care System of telemedicine, the medical information of a patient needs to be collected and safeguarded from the adversaries. This is attained by an efficient protocol which would provide access to the authorized persons to receive and update the medical information of a patient. In this paper, an efficient and secure Group Key Agreement protocol is designed which enables the authorized persons to agree on a Group key to access the private information and modify them. The designed protocol applies a Number-Theoretic function and an Affine operator to achieve robustness and security which is applicable for the smooth functioning of an e-Health Care System.
Keywords:	Cryptology, Group Key Agreement Protocol, Number-Theoretic Function, Affine Operator, e-Health Care System, Telemedicine
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	IMPROVED DEEP LEARNING SENTIMENT ANALYSIS FOR ARABIC
Author:	AHMED BINMAHFOUDH
Abstract:	Sentiment Analysis (SA) has recently gained great interest in Natural Language Processing (NLP). In fact, NLP consists in extracting data from texts and categorizing certain tweets as Positive, Negative, or Neutral. In this paper, we also present our participation in the Arabic Sentiment Analysis Challenge organized by King Abdullah University of Science and Technology (KAUST). Data of interest are tweets written in Arabic language, which becomes more challengeable. In this manuscript, we present the introduced system and the bi-LSTM model. Also, detail the less efficient explored solutions. Our main objective is to extract the crucial semantic data in Arabic tweets. The obtained findings about Arabic twitter corpus reveal that the performance of the developed technique is better than that proposed in the literature. Official test accuracy scores are 0.7605 with Macro-F1 score.
Keywords:	Attention, GRU, LSTM, Neural Network, Author Profiling, Gender Identiﬁcation, Deep Learning
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	AN IDENTIFICATION OF THE PROMINENT LEARNER BEHAVIORAL FEATURES TO PREDICT MOOC DROPOUTS USING HYBRID ALGORITHM
Author:	S. NITHYA, Dr S.UMARANI
Abstract:	Knowledge is derived throughout the world, which has launched plenty of online courses in response to the crisis. A growing number of learners are ready to enroll in online programs that help them adapt to this dynamic place. However, in some circumstances, the learner's level of retention rate should be reduced. The findings of a recent systematic review add to the many patterns that have been found in substantial research on learner digital interaction. This work used a set of distinctive characteristics derived from real-time datasets to measure learner dropout. The learning analytics framework is frequently used to examine user contributions and performance. Additionally, it supports higher education students' decision-making. To identify a learner's dropout, many researchers apply a variety of techniques, such as machine learning approaches and statistical methods. In this work, we propose Machine Learning-Artificial Intelligence (ML-AI), a novel hybrid approach that combines Random Forest (RF) and Artificial Neural Network (ANN) that requires minimal recurrent training to overcome these issues. The random forest is made up of a number of decision trees that can be used to classify data. It also assigns higher weights to the specified features in order to improve their classification capabilities. Then optimize by connecting the random forest to an ANN based on the feature score to obtain consistent prediction performance. Then we compared the metric values with the hybrid approach and some baseline models like Logistic Regression (LR) and Support Vector Machine (SVM). Our attempts are focused exclusively on diminishing early dropout rates and raising learner retention.
Keywords:	Dropout Prediction, Educational Data mining, Performance Evaluation, Artificial Neural Network, Multilayer Perceptron, Machine Learning
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	A LIGHTWEIGHT SPATIAL DOMAIN IMAGE ENCRYPTION ALGORITHMS: A REVIEW PAPER
Author:	ISSA JACAMAN, MOUSA FARAJALLAH
Abstract:	In our modern era, multimedia images have increased exponentially. Such data is being stored, transmitted through the public internet, and being used or produced by limited resource devices such as smartphones, Internet of Things devices, and healthcare devices. The conventional algorithms fail to protect such data while being processed by limited resource devices as it requires the most computational cost and increases communication overhead. In this short review, several lightweight encryption algorithms that overcome the conventional algorithm are considered. Moreover, we highlight some of the existing algorithms that are involved in encrypting images in the spatial domain for resource-limited devices and provided a comparison among them for their performance and robustness. The presented algorithms’ techniques were categorized for better understanding of the five categories in-compression approach, coefficient correlation, edge detection, most significant bits, and byte stream with a specific encryption ratio.
Keywords:	Chaotic Encryption, Selective Encryption, Cellular Automata, Lightweight Encryption, Stenography.
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	A HEURISTIC RANKING OF DIFFERENT SIMILARITY TECHNIQUES USED FOR EFFECTIVE LANGUAGE TRANSLATION AND PLAGIARISM DETECTION
Author:	PELURU JANARDHANA RAO, Dr. KUNJAM NAGESWARA RAO, Dr. SITARATNAM GOKURUBOYINA
Abstract:	With the rapid growth of language translation tools and digital libraries, text documents can be easily translated from one language to other resulting cross-language or multilingual plagiarism. Through this article we are presenting a detailed study in the comparison of multilingual documents for the efficient language translation. Parallel corpus is used to compare multilingual text which is a collection of similar sentences and sentences which are translation of each other. A detailed study is presented in this paper with various methods used in the literature to identify the similarity between French and English languages. A heuristic ranking model was developed to assess the suitability of various string similarity methods for determining language similarity. Through this study we concluded that the Fuzzy-Wuzzy (Partial-ratio) string similarity method out performs in terms of accuracy and Spacy similarity technique finds the similarity between the languages used for translation in less amount of time.
Keywords:	Natural Language Processing, Plagiarism, String Similarity, Fuzzy-Wuzzy, Sequence Matcher, Levenshtein Distance
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	AN EFFICIENT META-HEURISTIC OPTIMIZATION-BASED COMMUNITY CLUSTERING AND PRIVACY PRESERVING FRAMEWORK FOR LARGE OSN DATA
Author:	SHAMILA. M, G. REKHA, K. VINUTHNA
Abstract:	Privacy preserving plays an essential role to protect the patterns in the large online social networking databases. Most of the conventional community clustering and meta-heuristic models use static optimization functions in order to find the relationship among the local and global graph nodes for privacy preserving process. Also, these approaches use limited number of social networking graph nodes for community clustering and privacy preserving process. In this work, a meta-heuristic optimization-based community clustering framework is proposed to optimize the privacy preserving process. In this work, a homomorphic encryption-based privacy preserving approach is implemented to protect key nodes during the community clustering process. In this research, the cluster error rate of the inter and intra variations are minimized to improve the community detection process. The performance of the proposed community detection approach is tested on different online social networking datasets for community detection. Experimental results prove that the proposed meta-heuristic-based community clustering and privacy preserving framework has better efficiency than the conventional meta-heuristic-based privacy preserving methods on different OSN datasets. Finally, the privacy of the different social networking nodes and its properties are preserved on large number of graph nodes.
Keywords:	Online Social Networking, Privacy Preserving, Local Optimization, Global Optimization, Machine Learning.
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	THE ACCURACY OF MAPE MODEL ON CNN TO CLASSIFICATION OF MOVING OBJECT
Author:	AL-KHOWARIZMI, FATMA SARI HUTAGALUNG, HALIM MAULANA
Abstract:	Neural Network which is the adoption of the performance of the human brain based on various types of data that process into information and knowledge. There are many algorithms in neural networks such as Convolutional Neural Network (CNN). CNN is an algorithm that is able to provide knowledge based on image data and moving images. For example, moving images are traffic activities consisting of various vehicles and vehicle numbers. So there needs to be a classification in recognizing vehicle numbers from moving media for data center purposes in smart cities. Success in classification is of course the value of accuracy. However, CNN, which generally uses ordinary accuracy techniques, needs to be tested with other accuracy techniques such as Mean Absolute Percentage Error (MAPE) in order to obtain optimal accuracy models. From the accuracy model based on the usual formula, the highest 92% results from frame 2 while the MAPE accuracy model achieves 100% the highest from frame 4. And the last, his shows that CNN will be more optimal in moving image classification with the MAPE accuracy model.
Keywords:	CNN, MAPE, Moving Object, Vehicle Plate.
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	DIGITAL RISK ASSESSMENT AND PREDICTION IN TECHNOLOGY PROCESS STAGES OF ORE-STREAMS
Author:	ALMAS MUKHTARKHANULY SOLTAN, BAKYTZHAN TURMYSHEVICH KOPZHASSAROV, SAULE BELGINOVA, YURIY ANDREYEVICH VAIS, ZHANERKE KYDYRBEKOVNA AZAMATOVA, ZARINA TOLEUBEKOVNA KHASSENOVA
Abstract:	The goal of the work is to improve the quality of the management process by quantifying and predicting management and decision-making risks in multi-parameter systems, using ore streams as an example. The key to the ore management system in the paper is the control process. The control process appears to be a complex procedure involving measurement procedures, comparison of the measured value with the standard and decision-making. Under conditions of parametric uncertainty of control agents, control is accompanied by decision-making errors in the form of false and undetected rejects. In the true context, probable errors are defined as two types of risk: manufacturer's risk and client's risk. To quantify these risks, probabilistic and simulation models have been developed to investigate the impact of statistical characteristics of control agents and simulations on control outcomes. The validity and effectiveness of the modelling is tested with a computer experiment. The modelling approach developed is universal and can be used in a variety of scientific and technical practical applications. The authors have proposed a new multi-approach methodology for quantifying decision-making risks in a multi-parameter control system.
Keywords:	Management, Agent, Probability, Control, Credibility, Model, Norm, Error, Process, Risks, System.
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	SYSTEMATIC REVIEW OF THE ARABIC NATURAL LANGUAGE PROCESSING: CHALLENGES, TECHNIQUES AND NEW TRENDS
Author:	GHIZLANE BOURAHOUAT, MANAR ABOUREZQ, NAJIMA DAOUDI
Abstract:	The volume of Arabic posts on many social networks has increased signiﬁcantly, providing a rich source for analysis. As a result, Arabic Natural Language Processing intervenes to exploit this source and extract invisible but valuable insights. This paper presents a review of recent studies on techniques used in the Arabic Natural Language Processing field to come up with the faced challenges and the new trends. The articles selected for the review are primarily studies on Arabic Natural Language Processing techniques, as we collected and analysed a set of journal papers published in this field between 2018 and 2022. Based on the analysis, we extracted the various ANLP steps and investigated the techniques used in each step. The article also outlines the current trends in the several phases and steps of the Arabic Natural Language Processing process. As a result, it gives an insight into the current state of research.
Keywords:	Arabic Natural Language Processing, Systematic Literature Review, Data Collection, Tokenisation, Embedding.
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	A FRAMEWORK TO BUILD AND CLEAN MULTI-LANGUAGE TEXT CORPUS FOR EMOTION DETECTION USING MACHINE LEARNING
Author:	K ANUSHA, D. VASUMATHI, PRABHAT MITTAL
Abstract:	In the recent times, extraction of emotions from the text corpuses have gained a huge popularity. The use of these extracted emotions is used for various purposes such as customer reviews for products, recommendations of books, movies or building opinion poll from the social media posts. A good number of research outcomes can be observed during the last few years. Nonetheless, these existing systems have significantly failed to solve particularly two major challenges. Firstly, the text corpus available in various sources are in diversified languages and during the extraction of the emotions these variations of language can be highly complete to extract the correct emotions due to the dependency on regional languages. Secondly, the increase of using emojis in the text corpuses have made the task of emotion extraction even challenging since, the use of emojis with text can sometimes reflect to a sarcasm and detection of the sarcasm can be highly complex. Henceforth, this work proposes an automated framework to build a sentiment extraction process with pre-processing of the text corpus for normalization of the text from emojis, sarcasm and multiple local language influences. The framework results into 10% decrease in time complexity and 80% improvement over the accuracy of emotion detection using proposed machine learning methods.
Keywords:	Topic Inference, Text Corpus Extraction, Emoji Replacement, Emoji Translation, Text Translation, Mutual Exclusion, Emotion Extraction
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	MODEL CONSTRUCTION AND DEVELOPMENT OF AN ALGORITHM FOR THE PROBLEM OF PROPAGATION OF FLEXIBLE WAVES IN A HALF-PLANE WITH PARTIAL ABOUT AN ELASTIC STRIP PLACED
Author:	ZHUZBAYEV S., KHABDOLDA B., MURATKHAN R., TAKUADINA A., TANIRBERGENOV A., ABDYKHALYK A., TASZHUREKOVA ZH., ZHUMABEKOV M., SEITZHAN N.
Abstract:	The article considers numerical solutions of some spatial non-stationary problems for elastic and elastoplastic bodies of finite dimensions in the form of a parallelepiped, and for them the laws of propagation of three-dimensional waves are studied. An explicit difference scheme based on a combination of the methods of bicharacteristics and splitting in spatial variables is presented. Based on the described method, the elastic problem of longitudinal and transverse impact on a parallelepiped with one rigidly fixed end is solved. The features of the propagation of three-dimensional waves and the influence of a change in the speed of an external load on the pattern of wave propagation are studied, and some features of the propagation of dynamic stresses in the vicinity of a rigidly fixed end are revealed. The software was developed and tested for the robustness of the numerical calculation. Results, achievements of calculations in the form of isolines of normal stresses, performed for different points in time
Keywords:	Model, Algorithm, Flexible Waves, Half Plane, Elastic Strip
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	A NOVEL APPROACH FOR QR CODE IMAGE BASED ON DECIMAL CODING TECHNIQUE, GRAY LEVEL CO-OCCURRENCE MATRIX AND LOCAL BINARY PATTERN
Author:	TOUFIK DATSI, KHALID AZNAG, AHMED EL OIRRAK
Abstract:	Image description is an important task in image processing. Current image description techniques are based on grayscale and color image. In this paper, we propose an efficient approach for a QR code image description that combines a decimal coding technique with a gray-level co-occurrence matrix (GLCM) and a local binary pattern (LBP). This approach involves extracting the characteristics of the initial image by transforming the binary representation into a decimal value, i.e., by encoding each row of the image as a decimal value. All the values calculated from the image are regrouped into a matrix, which is used to extract a gray-level co-occurrence matrix and local binary pattern. The proposed method has been implemented using QR code dataset. Experiments show that the proposed approach effectively describes the QR code images and gives the best results in terms of GLCM properties, the histogram of the LBP, and the execution time.
Keywords:	Texture Analysis, Image Description, Decimal Coding, Gray-Level Co-occurrence Matrix, Local Binary Pattern
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	DETECTING AUTISM OF EXAMINEE IN AUTOMATED ONLINE PROCTORING USING EYE-TRACKING
Author:	TEJASWI POTLURI, DR.VENKATARAMA PHANI KUMAR SISTLA, DR.K.V.KRISHNA KISHORE
Abstract:	The purpose of this study is to alert the proctor in the automated Online Proctoring about the autistic examinee to closely monitor them. The existing studies have detected the autistic behaviour using MRI scan reports. In our work, the autistic behavior is captured through eye-tracking by generating heatmaps using hourglass modules. Eye movements of the examinee are recorded by calculating pith, yaw and roll angles on heatmaps generated by hourglass. Eye movements were tracked while they look for the information on the web pages. Eye movements data is used to train machine learning classifiers to detect the condition.. The autistic behavior is detected by using different gaze based tasks like search task, browse task and synthesis task. Experiments have also shown that eye-tracking data can be used for automatic detection of autistic examinee with the accuracy of 0.73. Results have proven that automated online proctoring system performs better with detecting autistic behaviour by using eye tracking without using MRI images.
Keywords:	Autism Detection, Gaze Tracking, Web, Automated Online Proctoring
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	AUTOMATION OF TEST CASE PRIORITIZATION: A SYSTEMATIC LITERATURE REVIEW AND CURRENT TRENDS
Author:	JIJO JOSEPH C GEORGE, D. PETER AUGUSTINE
Abstract:	An Important stage in software testing is designing a test suite [18]. The test case repository consists of a large number of test cases. However, only a portion of these test cases would be relevant and can find bugs. Test case prioritization(TCP) is one such technique that can substantially increase the cost-effectiveness of the testing activity. Using test case prioritization, more relevant test cases can be captured and tested in the earlier stages of the testing phase. The objective of the study is to understand different techniques used and a systemic study on the effectiveness of these approaches. The Literature consists of a few relevant articles introducing novel techniques for test case prioritization between 2008 and 2022. Studies show that parameters that are considered for test case prioritization are important. Hence, the current article also focuses on the parameters considered in the literature. 40% of the articles used in the literature review use different test case information as parameters. A systemic review and analysis of data sets involved in the literature are evaluated in the study. The review also focuses on the different approaches used for comparing the newly introduced approach and reveals a novel approach for prioritization.
Keywords:	Software Testing, Software Test Automation, Software Engineering, Test Case Prioritization, Regression Testing
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	THE EFFECT OF TECHNOLOGY ADOPTION MODEL ON CULTURALLY MEDIATED BEHAVIORAL INTENTION AND RISK PERCEPTION ON THE PURCHASE OF ELECTRONIC AND HOME LIVING PRODUCTS
Author:	DONG JIRUI
Abstract:	The purpose of this study was to determine the effect of the Technology Adoption Model on the culturally mediated behavioral intention and risk perception on the purchase of electronic and home living products. This research method uses a quantitative approach that is processed using SEM. The results of this study indicate that Perceived ease of use has a positive effect on Perceived Usefulness, Perceived ease of use has a positive effect on Perceived Usefulness, Perceived ease of use has a positive effect on Perceived Usefulness, Culture is not successful in mediating the relationship of Perceived Usefulness to Risk perception, Culture mediates the relationship Perceived ease of use on Risk perception, Behavioral Intention has a positive effect on Risk perception, and Culture mediates the relationship between Behavioral Intention and Risk perception. The limitations of this study include the sampling technique carried out by the convenience technique, namely the sampling is done as is. Data is collected over a short period of time so it is less likely to see consistency over a longer period.
Keywords:	Perceived Usefulness, Perceived ease of use, Culture, Risk perception, Behavioral Intention
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	A HIERARCHICAL ROUTING TREE TOPOLOGY BASED ON PROBABILISTIC SELECTION OF CLUSTER HEAD AND SUPER CLUSTER HEAD TO OVERCOME INTER-CLUSTER AND INTRA-CLUSTER ROUTING PROBLEMS IN DTN NETWORKS
Author:	EL MASTAPHA SAMMOU
Abstract:	DTN networks are environments that are characterized by a dynamic topology and which can change randomly and unpredictably due to high node mobility, frequent node disconnections, low node density and energy failures. All these factors create a set of routing problems namely: a very long delivery delay, a very low delivery rate and a significant increase in the resources consumed in the network. With all these problems, the hierarchical topologies proposed by the researchers are ineffective in terms of coordinating communications and data exchanges between the nodes of the different network clusters and are characterized by the absence of any kind of forecast against the premature death of cluster-Heads (CHs) and message ferries (MF). Therefore, it is necessary to develop hierarchical routing topologies adapted to routing problems in DTN networks to encourage the cooperation of nodes in the same cluster and also to coordinate communications between different clusters in the network. In this article, we will discuss routing improvement in DTNs by exploiting three factors: (a) hierarchical routing topology, (b) the probabilistic election of Cluster Head (CH) and the optimal election of Super Cluster Head (SCH), (c) inter-cluster and intra-cluster routing. We propose a hierarchical routing tree topology (HRTT) based on a probabilistic approach for the election of Cluster Head (CH) and Super Cluster Head (SCH) on the basis of three criteria: The probability of contact of the Message Ferry (MF), the residual energy and the node mobility model. The results of the simulations carried out show the efficiency of the topology (HRTT) with a considerable increase in the delivery rate and a minimization of the average latency in the DTN networks.
Keywords:	Delay Tolerant Networks (DTN), Tree Topology, HRTT, Super Cluster-Head, Cluster Head Election, Hierarchical Routing, ONE Simulator.
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text

Title:	NEW APPROACH TO AUTOMATIC AND RELEVANCE DECISION-MAKING IN A SMART HOME BY COMBINING A HIGH-LEVEL CONTEXT-MANAGING ONTOLOGY AND A RULES BASE
Author:	MOHAMED EL HAMDOUNI, YASSER MESMOUDI, ABDERRAHIM TAHIRI
Abstract:	The Smart Home is a residence equipped with computer technology that assists its inhabitants in the various situations of domestic life by trying to optimally manage their comfort and safety by action on the house. The detection of abnormal situations is one of the essential points of a home monitoring system. These situations can be detected by analyzing the primitives generated by the audio processing floors and by the apartments sensors. We propose in this paper a Inference and decision-making model based on a high-level ontology and rules base.
Keywords:	IA, Decision, Ontology, Smart Home, Rules.
Source:	Journal of Theoretical and Applied Information Technology 15^th February 2023 -- Vol. 101. No. 3-- 2023
Full Text