		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 2022 | Vol. 100 No.17

Title:	USER EXPERIENCE PERCEPTION OF ANDROID-BASED THERAPIST SPA RESERVATION APPLICATION (A CASE STUDY OF BABY SPA APPLICATION USERS)
Author:	CARUDIN, UWAY WARIAH, ASEP JAMALUDIN, NELLY APRININGRUM, DADANG YUSUP
Abstract:	The increasing public demand for homecare baby spa services has made business actors in this field start trying to compete online in providing the best services and services. Efforts are made especially in the use of information technology for the context of media marketing, one of which is using application media. Applications that are currently believed to be able to reach a broad market so that they require more attention, especially in terms of design and content, this is because design and content can provide a user experience so that they are interested in using the application. This study is to identify the application of the User Experience (UX) concept to the baby spa therapist reservation application, so that it can be a reference for future developers who want to develop similar mobile applications. Statements for respondents are based on 4 fundamental UX variables, namely Value, Adoptability, Desirability and Usability which are made on a Likert Scale. Statistical methods used are validity and reliability tests. Furthermore, the results of the questionnaire are presented in the form of descriptive analysis based on the four UX variables. Meanwhile, to find out the most important fundamental UX elements, the "YES & NO" classification method is used. Based on the research results, it is known that the user experience variables that are most considered important by respondents are the value, adoptability and desirability variables. This is because people use this baby spa application to reserve a baby spa therapist, it is very easy to learn, has an attractive visual design and is easy to access.
Keywords:	Baby Spa, User Experience, Information Technology, Fundamental UX Variables
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	CHINESE CHARACTER RECOGNITION USING SUPPORT VECTOR MACHINE
Author:	ASLINA BAHARUM, ROZITA ISMAIL, SHALIZA HAYATI A. WAHAB, FARHANA DIANA DERIS, NOORSIDI AIZUDDIN MAT NOOR, MOHD SHAREDUWAN MOHD KASIHMUDDIN
Abstract:	Optical character recognition is the art of scanning and detecting the word in the images so that the machine can identify and classify the character. Chinese characters are one of the world's most widely used writing systems. It is used by more than one-quarter of the world’s population in daily communication. Chinese characters can be considered difficult because they have many categories, complex character structures, similarities between characters, and various fonts or writing styles. There are many known machine learning algorithms for character recognition, but not all can classify Chinese characters with high speed and accuracy. Therefore, this paper proposes recognizing Chinese characters using support vector machines. Support vector machines are a classification of two classes widely used in classification. It produces very accurate results for many classes, making it suitable for recognizing Chinese characters.
Keywords:	Support Vector Machine, Optical Character Recognition (OCR), Character, Classifier, Feature
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	EXPERIMENTAL INVESTIGATIONS TO AUTHENTICATION OF PERSON USING HAND RADIOGRAPHS
Author:	RAM GOPAL MUSUNURU, Dr. T SIVAPRAKASAM, Dr G KRISHNA KISHORE, Dr. R THIRUVENGATANADHAN
Abstract:	Confirmation is the most common way of distinguishing client's character. The personalities that are given by client are contrasted with those on a record in a data set of the approved client's data on a neighborhood working framework or inside a validation worker. Biometric radiographs are utilized in confirmation of an individual during wrongdoing and debacle episodes. As of late, confirmation and recognizable proof of an individual has turned into a significant piece of the vast majority of the mechanization frameworks. The different biometric distinguishing pieces of proof like unique finger impression, iris, face, palm motifs neglect to perceive the human when the outer biometric parts have been harmed because of epidemics, lesions, and serious consuming. Securities, heartiness, protection, in addition non-imitation are the basic parts of any individual confirmation framework. In such circumstances, recognizable proof dependent on radiographs of the skull, hand, and teeth remain viable substitution techniques. In our examination, an original legal hand radiograph centered human validation is offered utilizing a profound neural organization. We use convolutional profound neural organization (CNN) design intended for the component taking out of hand radiographs and for acknowledgment. As a piece of this we have taken 1400 unique kinds of hand radiographs dependent on their age gatherings, callings, and sexes are thought of. Our investigation uncovers that hand radiographs encompass biometric data that jerry can be utilized in the direction of recognize people in misfortune casualty ID. The trial concentrate likewise shows that the proposed approach is altogether powerful than traditional strategies for the individual confirmation utilizing hand radiographs. In this we use 4 layers, initially the input image is passed through convolution layer and after processing it is passed through relu layer to remove the negative values. The output from this layer is passed to pooling layer where max/ average pooling be situated applied on it. And finally passed to fully connected layer where the final classification be there done and determines whether the hand radiograph are valid or invalid.
Keywords:	Authentication, Hand Radiographs, CNN, Investigation, Biometric
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	OPTICAL HEVC CRYPTOSYSTEM USING INPUT AND FRACTIONAL FOURIER PLANES RANDOM ENCODING
Author:	MOHAMMED A. ALZAIN
Abstract:	This paper present an efficient optical High efficiency video coding (HEVC) cryptosystem using input and Fractional Fourier (FrFT) planes random encoding. The encryption of the proposed optical HEVC cryptosystem using input and FrFT planes random encoding starts by separating each HECV plainvideo frame into RGB channel components and every one of HEVC RGB channel components is modulated in the input plane with the first random phase mask (RPM1). After that, the FrFT is applied to each one of HEVC RGB channel components. Then, every one of modulated HEVC RGB channel components in the FrFT is modulated again with the other RPM2 and the inverse of FrFT is employed. Finally, the HEVC RGB channel components are merged to get the HECV ciphervideo frame. The decryption of the proposed optical HEVC cryptosystem using input and FrFT planes random encoding starts by separating each HECV ciphervideo frame into RGB channel components and every one of HEVC ciphervideo frame RGB channel components is subjected to FrFT and modulated in the FrFT plane using the conjugate of RPM2. After that, the inverse of FrFT is applied to every one of modulated HEVC ciphervideo frame RGB channel components and modulated in the input plane using the conjugate of RPM1. Finally, the decrypted HEVC ciphervideo frame RGB channel components are combined to get the deciphered HECV frame. The proposed the proposed optical HEVC cryptosystem using input and FrFT planes random encoding is tested with different security metrics like visual inspecting, statistical, cipher quality, differential, and occluding tests. The outcome of tests ensures and confirms the effectiveness of the proposed optical HEVC cryptosystem using input and FrFT planes random encoding.
Keywords:	Optical Encryption, HEVC, Fractional Fourier Transform (FrFT)
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	EFFECTIVE DATA ALIGNMENT AND KEY EXTRACTION TECHNIQUES FOR GENERATION OF REGULAR EXPRESSIONS USING MACHINE LEARNING TECHNIQUES
Author:	DINESH D. PURI, Dr. G. K. PATNAIK
Abstract:	Machine learning supervised classification plays a significant role in large text classification. Health care data contribute to generating privacy and security for a couple of years. Such electric records might take extensive data from storage devices, so it needs to optimize with some processing techniques. For generation of regular expression, data should be in structured form. The conversion of unstructured data in to structured form is done through modified data alignment and key extraction techniques. The Smith-Waterman method compares two sequences to find comparable review statements. Smith-Waterman method employs an evolutionary algorithm to find improved local alignments of pairs. It can identify the best local alignment using the supplied score system. The NLP process has been used to generate the keys from such large text. In this paper, we proposed a sequence alignment generation using Smith-Waterman (SW) algorithm and key extraction from generated sequence using Natural Language Processing (NLP) technique has used for effective regular expression building. The filtration techniques have been used to eliminate redundant features, and the Machine Learning (ML) algorithm has been used as post-processing for classification. The generated regular expression by the SW algorithm gives better classification accuracy using NLP and Machine leavening. Generated regular expressions are filtered using 100% precision as threshold. We applied various Machine learning algorithms out of which SVM gives highest accuracy.
Keywords:	Sequence Generation, Sentence Alignment, Key Extraction, Regular Expression, Classification, Smith Waterman Algorithm, NLP
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	HIGH SPEED DECODING BY COLLABORATION BETWEEN THE HARTMANN RUDOLPH AND INFORMATION SET DECODING ALGORITHMS
Author:	HAMZA FAHAM, MY SEDDIQ EL KASMI ALAOUI, SAÏD NOUH, MOHAMED AZZOUAZI, ISSAM ABDERRAHMANE JOUNDAN
Abstract:	Decoders are implemented to retrieve data after its transmission over a noisy communication channel. Soft decision decoders are highly efficient in concatenation schemes exploiting more than one decoding level. In our case, we concatenated the symbol-by-symbol Hartmann Rudolph (HR) decoding algorithm and the Information Set Decoding (ISD) technique that is a word-to-word decoding. In this work, we will suggest to concatenate HR partially exploited (PHR) and the ISD technique in order to decode linear block codes. We will use firstly the HR decoder with a reduced number of dual codewords then the ISD, which uses the output of PHR. We noticed that the suggested serial concatenation guarantees very high performances with less dual codewords number. For instance for the QR(31, 16, 7) code, the satisfying obtained results are based only on 2,74% of the dual codewords. For the same code, we have minimized the runtime by 95% compared to the use of HR alone. This proves the power and the speed of the suggested concatenation.
Keywords:	Information Theory, Error Correcting Codes, Hartmann Rudolph (HR), Information Set Decoding, PHR
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	CLASSIFICATION MODELS COMPARISON FOR PARKINSONS DISEASE DETECTION THROUGH SPEECH FEATURES
Author:	CHRISTOPHER PUTRA SETIAWAN, EDWARD PRATAMA PUTRA, GEDE PUTRA KUSUMA
Abstract:	Prior researches have used speech features to classify Parkinson’s disease. This research was conducted to compare thirteen classification models in order to achieve a better performance in the newly published dataset. The method offered in this paper involves the usage of hyperparameter tuning, feature selection using genetic algorithm, and model stacking with ANN as the final classifier. The final stacked model has achieved an accuracy of 90.13% in the test dataset, with an F1- score of 93.7%. The results indicate that classification accuracy of Parkinson’s disease classification through speech features can be increased by utilizing hyperparameter tuning, feature selection, and ensemble stacking. This improvement also indicated that the usage of feature selection in tandem with ensemble stacking method can yield better result compared to prior state of the art models for Parkinsons disease detection.
Keywords:	Parkinsons Disease Detection, Machine Learning, Ensemble Stacking, Hyperparameter Tuning, Feature Selection
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	A PERFORMANCE OPTIMIZED REVERSIBLE CELLULAR AUTOMATA BASED SECURITY ALGORITHM FOR SECURE DATA COMMUNICATION AND STORAGE IN HEALTHCARE
Author:	SURENDRA KUMAR NANDA, SUNEETA MOHANTY, PRASANT KUMAR PATTNAIK
Abstract:	Information security and protection of privacy is one of the critical tasks in any field of work. It is more critical for the field of medical science data as it involves life and critical healthcare hazards. Another aspect is the advances in technology that gives high processing capabilities to the users and they can use these high capabilities system to protect this information more effectively and efficiently. The privacy of information should be ensured not only on data stored in storage but also required during electronic communication from one place to other. The reversible cellular automata and cryptographic hash functions can be used to achieve security of data on rest and during the transition with optimal performance. We used a 128 bits encryption algorithm based on reversible cellular automation to protect information during transit and 128 bits cryptographic hash function to protect data at rest. The implementation results show these algorithms are immune to cryptanalytics and extremely difficult to break it using brute force attack. The NIST statistical results show good characteristics of this algorithm. The hardware implementation of these algorithms is simple as it uses simple logic gates and hence cost of implementation is cost-effective
Keywords:	Reversible Cellular Automata, Cryptographic Hash Function, Clinical Data Security, Healthcare
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	A NOVEL APPROACH FOR CONTENT EQUIVALENCE ANALYSIS IN COMPRESSED DOCUMENT IMAGES: A SYSTEMATIC STUDY
Author:	KAVITA V. HORADI, DR. JAGADEESH PUJARI, NARASIMHA PRASAD BHAT
Abstract:	Rapid growth of digital data with complex content has led to various challenges in processing. Exponential increase in the size of ‘Big Data’ due to videos, audios, images and textual content has created several problems which need to be addressed by the research community. Currently, huge amount of digital data is generated by various sources. The high quality data require more space and consume excessive bandwidth during transmission. To overcome these issues, digital data are stored in compressed form using different compression algorithms stated in literature. In order to analyze these data traditional schemes use decompression techniques which are a time consuming process and increases the computation overhead. To overcome these issues, currently compressed domain image processing techniques have been adopted where complete decompression may not be required. In this work, we adopt document image processing in compressed domain which contains printed text in the document images. Our main aim is to identify the similarity and find the equivalence between two or more compressed document images. In order to achieve this, first of all, we apply JPEG encoding which generates encoded data. This data further processed through the proposed line, word and character segmentation scheme. Further, we apply SIFT (Scale-Invariant Feature Transform) to extract the feature from compressed domain segmented data. Finally, feature matching scheme is applied which uses Brute force feature matcher and k-nearest neighbor. We have tested this approach on publically available PubLayNet, IIIT-AR-13K, and Tobacco-3482 datasets which contains large scale document images. The experimental analysis shows the robustness of proposed approach to identify the similarity between compressed documents images.
Keywords:	JPEG; Compressed Document Images (CDI); Document processing; Content Equivalence; Compressed Domain; SIFT; Brute force; KNN.
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	TELUGU TEXT SUMMARIZATION USING HISTO FUZZY C-MEANS AND MEDIAN SUPPORT BASED GRASSHOPPER OPTIMIZATION ALGORITHM (MSGOA)
Author:	CHINNI BALA VIJAYA DURGA, DR. G.RAMA MOHAN BABU
Abstract:	In this work, we clearly stated the text summarization. Time is taken to read the big document it's quite complex, at the same time if we summarize the same document it is easy to understand and time-saving. Here we propose the text summarization of the Telugu language. To begin, documents must go through a preprocessing procedure that includes tokenization, stop-word removal, stemming and N-gram analysis. After that, it uses Histo-Fuzzy C-means Clustering to achieve clustering, as well as a technique of sentence ranking based on weights. Finally, the Median Support Based Grasshopper Optimization Algorithm (MSGOA) is utilized to combine the phrases into a clear and succinct summary. The performance of this strategy is evaluated using an online research dataset. When compared to earlier text summarizing methods, the suggested method outperforms them. When compared to existing accuracy, the proposed method performs admirably and obtains an accuracy of 84%.
Keywords:	Telugu language Text Summarization, preprocessing, Median Support Based Grasshopper Optimization (MSGO), Histo-Fuzzy C-Means Clustering, Stemming, Enthalpy
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	DESIGN OF A CONCEPTUAL FRAMEWORK FOR A GAMIFIED SOCIAL MEDIA APPLICATION
Author:	OHWO ONOME BLAISE, SEUN EBIESUWA, ADESINA ADIO, ADEGBENJO ADERONKE, AMUSA AFOLARIN IBRAHIM
Abstract:	The goal of gamification is the deployment of game mechanics in real-world context for non-gaming scenarios, to drive motivation and performance regarding a given activity. Preceding research, mostly supports the hypothesis fundamental to this aim. Till date, gamification has continued to find its application in various industries and sectors. Among application area where gamification has found its usage are in social networks. Due to the commercialization of the mobile devices and user content generation, there has been the emergence of mobile social networking, which has changed the way the users relate to applications. In the search for better ways to improve user experience in mobile social networking, gamification (the use of game design elements in non-related games) can play a vital role. And considering there are many different game mechanics which can result in diverse application services. Various game mechanics were reviewed to understand the working principles and where they can be applied. Based on this review, a conceptual framework for gamified social media application was presented. The game mechanics configuration was deliberately varied, in regards to their impact on the completion of basic functionalities as well as potentially improve user engagement.
Keywords:	Gamification, Game Mechanics, Conceptual Framework, Social Media, Social Networking, User Engagement, Motivation
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	IWVTSA: IMPROVED WORDS VECTORS FOR TWITTER SENTIMENTS ANALYSIS
Author:	AMINA SAMIH, ABDERRAHIM GHADI, ABDELHADI FENNAN
Abstract:	Twitter sentiment analysis technology provides methods for polling public opinion on events or products. The majority of current research is aimed at obtaining sentiment features by analyzing lexical and syntactic features. Word embeddings express these characteristics explicitly. This paper proposes a new approach called Improved Words vectors for Twitter sentiments analysis (IWVTSA) to improve the f1 score of twitter sentiment analysis. We introduce a word embeddings method called tweet2vec, which was launched concurrently with doc2vec, to form a sentiment feature set of tweets; these word embeddings are combined with word sentiment polarity score features. For training and predicting sentiment classification labels, the feature set is integrated into four machine learning classifiers (XGBoost, SVM, Logistic Regression, Random Forest). We compare the quality of our model to that of the baseline model, the bag of words algorithm, and the results show that the combination of tweet2vec and Polarized lexicon in our model performs better on the F1-measure for Twitter sentiment classification.
Keywords:	Sentiment Analysis, Tweet2vec, Doc2vec, Polarized Lexicon, Machine learning.
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	DEEP LEARNING BASED SOFTWARE RELIABILITY PREDICTION MODEL FOR COMPONENT BASED SOFTWARE
Author:	SHIVANI YADAV, BALKISHAN
Abstract:	Component Based Software Engineering (CBSE) is a software development approach used for rapid development. This involves the use of existing, mature components, which might have been developed for other software or product, along with new components to speed up the software development process. Component based development is being increasingly adopted to reduce costs, manage the increasing complexity of software products, and faster time to market. A major challenge posed by the adoption of Component Based Software (CBS) is to ensure the reliability of software products developed using this approach. The reliability of CBS is dependent upon the individual component reliability as well as software architecture. In this paper, a deep learning-based model is designed to predict the reliability of components. The results show that the designed model is well predicting the reliability with high accuracy. Comparison with other machine learning models justified the use of deep learning models in reliability prediction.
Keywords:	CBS, Component, Deep Learning, Prediction, Reliability
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	DEEP CONVOLUTIONAL NEURAL NETWORK IMPLEMENTATIONS FOR EFFICIENT BRAIN STROKE DETECTION USING MRI SCANS
Author:	KASHI SAI PRASAD, DR. S PASUPATHY
Abstract:	Advancements in Artificial Intelligence (AI) has paved way for solving the problems in different domains. In healthcare industry, brain stroke incidence is in alarming rate across the globe. According to WHO, it is one of the leading causes of disability and depth which needs sustainable research effort. It is believed that bringing AI into Clinical Decision Support System (CDSS) has potential to have unprecedented strides in healthcare units for improving Quality of Service (QoS). The existing MRI scans based research with deep learning has limitations as it uses pre-defined CNN based models. The CNN models such as VGG16, ResNet50 and DenseNet121 have their built in functionalities. However, for a given problem, their performance may not be adequate and they are to be optimized to meet requirements of the problem in hand. To overcome this problem, in this paper, we proposed a framework and underlying mechanisms to optimize and exploit the deep CNN models aforementioned for efficient brain stroke detection. Apart from the improving the models, an algorithm named Optimized Deep Learning for Brain Stroke Detection (ODL-BSD) is proposed. MRI scans are used to evaluate the proposed framework. The empirical results revealed that the proposed methodology has caused improvement in the performance of the deep CNN models significantly.
Keywords:	Brain Stroke Detection, Deep Learning, Convolutional Neural Network, Densenet121, Vgg16, Resnet50
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	GENERIC FRAMEWORK FOR EARLY PREDICTION OF HEART DISEASE USING LEARNING METHOD
Author:	VENKATESWARARAOCHEEKATI, DR .D.NATARAJASIVAN , DR.S.INDRANEEL
Abstract:	The incidence of heart disease is growing at a startling rate, making identification at an earlier stage all the more important. The identification of heart illness is a challenging endeavour that calls for accuracy and dexterity on the part of the diagnostician. To determine which people, based on a variety of different medical criteria, are more likely to develop heart disease, the purpose of this study is to identify those patients. Models for determining the likelihood of a person being diagnosed with a cardiac disease have been developed, and these models can be used to evaluate an individual's risk.SGD Classifier, Additional Tree Classifier, Calibrated ClassifierCV, Gaussian Mixture, Nearest Centroid, MultinomialNB, Logistic RegressionCV, Linear SVC, Linear Discriminant Analysis, SGD Classifier, Calibrated ClassifierCV, Linear SVC, Linear Discriminant Analysis, SGD Classifier, Calibrated ClassifierCV, Quadratic Discriminant Analysis, GaussianNB, Random Forest Classifier, ComplementNB, MLP Class; Gaussian Mixture; Nearest Centroid;, LGBM Classifier, Ada Boost Classifier, A very helpful method was utilised in order to determine how the model might be applied to improve the level of accuracy with which heart disease can be predicted in any given individual. Using Deep Learning and Random Forest Classifiers, which had a high degree of accuracy compared to previous models, the suggested model was able to effectively predict a person's likelihood of acquiring heart disease. The algorithm for predicting cardiac disease that has been proposed would not only make medical care more effective but would also reduce expenses. By making use of this knowledge, we are able to improve our ability to anticipate who might develop heart disease in the future. The model was constructed in Python, and the data was taken from the Kaggle vault.
Keywords:	Disease of the Heart, Artificial Intelligence, Deep Learning, Prediction
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	TOWARDS A NEW CLASSIFICATION METHOD OF TEAM PERFORMANCE USING CLUSTERING AND FEATURES REDUCTION TECHNIQUES
Author:	ZBAKH MOURAD, AKNIN NOURA, CHRAYAH MOHAMED, ELKADIRI KAMAL EDDINE
Abstract:	The human factor is becoming more and more decisive in making the company more efficient and more competitive, improving the performance of their teams represents a challenge for the Human Resources department, which is also experiencing a profound digital transformation of data and its management [1]. Traditional Human Resources tools are no longer effective at managing skills, On the one hand, the performance evaluation of these resources objectively becomes more and more one of the very complex tasks of a manager, especially with the mass of current data presented by new non-traditional sources, on the other hand, the departure of key skills is a phenomenon that is not predictable by current HRIS tools, the financial cost is high as well as the technical loss of knowledge and know-how, and flexibility that this presents. In this article, we will propose a new approach to the classification of teams according to several performance indicators. This method is based on the K-means algorithm to classify the members of a team, assessed against performance indicators linked to some soft and technical skills. The result of this work represents a decision support model for managers to develop a team adapted to the overall mission, to adapt its management style to each cluster, and to prepare future hires to compensate for the skill gaps of the team in place.
Keywords:	Human Resources Management, Key Performance Indicators, K-Means, Cluster Algorithm
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	JAVA SOURCE CODE SIMILARITY DETECTION USING SIAMESE NETWORKS
Author:	DIVYA KUMARI TANKALA, Dr. T. VENUGOPAL, VIKAS B
Abstract:	Software plagiarism checkers can be important during coding competitions, to review, evaluate, and rank the participants. For a problem statement, if the number of submissions is relatively small then inspecting each code submission and being able to determine whether they are similar to the existing code segment or not is easy, but in the case of huge code submissions, it is very difficult to determine the presence of code clones in the submitted code snippet. Therefore, there is a need for plagiarism checkers to detect similar clones. In existing studies, we could use various approaches to detect code clones, but code clone detection using an abstract syntax tree is one of the popular approaches. Our proposed approach based on AST is experimented on a BigCloneBench dataset consisting of Java code fragments and implemented using recursive neural networks. The siamese networks were used to detect similarities between two code fragments and presented the influence of contrastive learning in source code clone detection with high accuracy. This paper showcases the improvement in precision, recall and F1-score at least with 5% compared to existing approaches.
Keywords:	Code similarity, Plagiarism detection, Siamese networks, Recursive Neural Networks, LSTM, Contrastive learning
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	OPTIMAL OMNIDIRECTIONAL OFFLOADING IN FEDERATED CLOUD-FOG SYSTEMS WITH COST MINIMIZATION
Author:	SAUMYARANJAN DASH, RAJ KUMAR PARIDA, JYOTIPRAKASH DASH, ASIF UDDIN KHAN, SANTOSH KUMAR SWAIN
Abstract:	Cloud computing technology has brought a significant impact on the field of technology. Similarly, fog computing provides similar facilities with a lesser coverage area but is closer to the user. In a federated environment, these technologies can complement each other by offloading their request to improve the services. In this paper, we propose a federated two-tier architecture consisting of cloud and fog, where the cloud has a higher cost and fog has a lower cost. In this architecture, omnidirectional offloading is considered in which offloading can be done vertically and horizontally, such as Fog to Cloud, Cloud to Fog, and Fog to Fog. We propose a capacity optimization problem in this federated architecture to optimize the capacities to minimize the total cost. We are using a modified simulated annealing algorithm to solve this optimization problem. The results show the performance of our algorithm is nearly optimal and better than the SA in terms of execution time, and our proposed architecture can save more cost than other existing architectures.
Keywords:	Cloud, Fog, Federation, Offloading, Cost, Capacity
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	A MACHINE LEARNING BASED FRAMEWORK FOR BALANCING USER PRIVACY AND UTILITY WITH MULTIPLE SENSITIVE ATTRIBUTES IN HEALTH CARE DATA PUBLISHING
Author:	JAYAPRADHA. J, PRAKASH. M
Abstract:	Due to the increase in the emergence of health care system information, patient records are needed for analyzing certain diseases. It leads to the need for privacy and several challenges in the current health care system. Privacy-preserving data publishing is essential to protect the patient's record from numerous attacks. The main aim of privacy-preserving data publishing is to safeguard an individuals’ personal information, though it is made available for various analysis purposes. Initially, the paper has analyzed different services of an electronic health record, need for privacy and has proposed a novel anonymization technique for electronic health records. The proposed approach will overcome the following drawbacks i) generalization of all attributes which significantly reduces the information ii) identity disclosure even with adversaries having intense background knowledge and iii) the trade-off between privacy and utility. Compared to the existing system the proposed approach achieves a better result by using hierarchical id-based generalization approach. Hence, the proposed approach helps significantly in protecting individuals’ information even though an intruder has intense background knowledge. Additionally, it focuses on achieving balanced utility in the anonymized data by avoiding over generalization. The proposed approach consists of four phases i) vertical partitioning, ii) Quasi-identifier bucket(QB) anonymization, iii) Sensitive attribute bucket(SAB) anonymization iv) Evaluation of classification accuracy. As per the experimental result, the proposed approach achieves improved data privacy and utility in privacy-preserving data publishing. The experimental results are not evaluated using the utility metrics. However, the results, evidently illustrate that the proposed algorithm achieves improved accuracy than the standard utility aware data anonymization algorithm in health care records. The proposed approach thwarts the identity disclosure and attribute disclosure for the static data.
Keywords:	Anonymization, Health Care, Generalization, Privacy, Utility, Classification Model.
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	MINIMIZE THE SPREADING OF FAKE NEWS AND FORMING FAKE USER GROUP
Author:	AKASH D. WAGHMARE, Dr. GIRISH. K. PATNAIK
Abstract:	Due to the widespread use of social media-based international news, confirmation and identification has become a difficult undertaking. Most social networking sites make it simple to obtain news from anywhere on the internet at any time, but they also create a lot of fake news and information at the same time. As a result, in such a situation, it is vital to establish whether the information given is authentic, that is, whether it is fake or real. This causes people to become perplexed and lose faith in social media. To address these issues, the suggested blockchain-based false news detection using machine learning is applied. In training and validation evolution, the suggested classification algorithm was applied to detect bogus news. Another significant goal of this project is to revoke the offenders who update the news that has already been published or fake in blockchain framework. Even in a vulnerable context, the blockchain-based decentralized peer-to-peer ecosystem was leveraged to protect the published data. To produce successful training rules and verify the test classifier, a variety of feature extraction and selection strategies were applied. On the LIAR dataset, an exhaustive experimental investigation reveals the prediction performance of detecting false news with hybrid classification algorithm. The 94.6% accuracy obtained for fake news detection as well as detection of fake users’ identification. The major outcome of this research it reduces the frequency of spreading the fake news using blockchain majority algorithm with trust building strategy for end user.
Keywords:	Fake news detection, Supervised machine learning, Blockchain, Peer to peer, Majority voting, Consensus
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	MARKETING COMMUNICATION STRATEGY OF GOLD MINING COMPANY “PT ANTAM LOGAM MULIA” THROUGH DIGITAL PLATFORM
Author:	OSYE MEIRIANA SORAYA, ULANI YUNUS, MAHAPUTRA DIMAS ADIPRADANA
Abstract:	Technological developments in the digital era have made social media a "BOOM" in the 21st century. Because more than 3 billion people in the world use digital platforms for social media applications, and it is estimated that the number will continue to increase every year. The presence of social media has dramatically changed the communication activities of all lines of people's lives in the world. Not only is it used for chatting (or) communication purposes between two individuals but social media can also be used as an excellent marketing tool. No wonder companies use social media as a marketing communication strategy. This study aims to determine the implementation of marketing communication strategies through digital platforms implemented by PT Antam Logam Mulia as one of the largest gold mining companies in Indonesia. This study uses a qualitative descriptive methodology. Researchers conducted in-depth interviews with several experts in Marketing Communication Strategy at PT Antam Logam Mulia and technical implementers, as competent resource persons in answering the problems to be studied. The results of the study reveal the marketing communication strategy model through digital platforms (social media) implemented by PT Antam Logam Mulia, that the main direction of the marketing communication strategy is to increase sales, focus on the goals and target market by using organic marketing and Ads marketing. This study has implications for promoting digital aspects that do not only appear in the communication mix, but the overall process of marketing communications carried out by PT Antam Logam Mulia using digital platforms - social media to carry out marketing activities such as disseminating information, influencing or persuading, and or reminding target market for the company and its products to be willing to accept, want to buy, and be loyal to the products offered by the company.
Keywords:	Marketing Communication, Digital Promotion, Social Media, Precious Metals Gold, Social Networking
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	IMPROVING LIGHTWEIGHT AUTHENTICATION USING NEW TECHNIQUES FOR IOT
Author:	WALEED KAREEM AHMED, RANA SAAD MOHAMMED
Abstract:	The Internet of Things (IoT) is a new paradigm that uses an Internet to communicate a wide range of physical objects with the cyber scientist. The Internet of Things (IoT) is rapidly growing and would soon have a big influence on our daily lives. While the growing amount of linked IoT gadgets makes our lives easier, it actually puts our personal information at danger. For IoT devices, radio frequency identification (RFID) aids in the automated identification of connected devices. However, both privacy and security for RFID tag-connected technologies are the key issues. The increasing security of radio frequency identification (RFID) solutions for a variety of RFID applications that require a centralized database expansion, as compared to a standard central database, blockchain technologies are rapidly establishing itself with a new decentralized and distributed alternative that offers improved data security , dependability, transparency, the immutability, and lower maintenance costs. RFID is expected to play a major role in enabling identification technologies in the Internet of Things due to its inherent benefits. However, because of its connection with sensor technology, it may be used in the broad range of sectors. On the other hand, one of the most challenging parts of developing a RFID system appeared for being security. Authentication and privacy concerns are at the heart of RFID security. Elliptic curve cryptosystem (ECC) related algorithms are commonly regarded as the best option among PKC approaches due to their small key sizes and effective calculations. Recently W.K. Ahmed et al . proposed a New Lightweight BLOCKCHAIN and ECC-Based RFID Authentication Protocol for IOT. We found the weaknesses of W.K. Ahmed et al protocol by computation cost high and running time high. In order to solve these problems. In this paper, we introduce Improving lightweight authentication using New techniques for IoT. Our protocol uses techniques from them blockchain, ECC, Arnold map chaotic, and Markova chain. We implemented our suggested programming using python language. after comparing storage cost, communication cost, and computation cost with other protocols, our protocol is more secure and performance efficient than the existing RFID protocols and is well suited for practical applications.
Keywords:	ECC, Arnold Map Chaotic, Blockchain Technique, Authentication, Markov Chain
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	HARDWARE IMPLEMENTATION METHOD OF SECRET DATA SECURITY ON FPGA BASED ON ZIG-ZAG MAP ENCRYPTION AND STEGANO ALGORITHMS
Author:	BAYU KUMORO YAKTI, SUNNY ARIEF SUDIRO, SARIFUDDIN MADENDA, AND SURYADI HARMANTO
Abstract:	The internet has grown so rapidly that most of the individuals prefer to use the internet as the main medium for transferring data. Data protection is very important especially when sending secret data from one place to another. This paper proposes a zig-zag mapping transposition encryption-decryption algorithms and Least Significant Bit (LSB) steganography algorithm for data security improvement. The hardware implementation methods of these algorithms into Field Programmable Gate Array (FPGA) for data protection in real-time communication are also offered. The FPGA Intellectual Property core (IP core) resulting employed minimal LUTs resources. The encryption and steganography algorithms occupy 107 LUTs and take 1.821 ns for each 64 bits data processing. Whereas steganalysis and decryption algorithms needed108 LUTs and 2.172 ns processing time.
Keywords:	Encryption-decryption, LSB steganography, LUTs FPGA resources, Zig-zag Mapping Transposition
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	OPTIMIZING SPEAKER RECOGNITION USING D-VECTOR AND X-VECTOR SPEECH EMBEDDINGS AND DEEP LEARNING
Author:	KEVIN KURNIAWAN, AMALIA ZAHRA
Abstract:	In speaker recognition, a way to identify individual’s identity is to identify the characteristics of the individual voice, which can be identified using speaker’s speech embedding. In this paper, we propose a method using x-vector and d-vector speech embedding feature extraction and ResNet50 model on identifying user’s identity. This paper uses part of VoxCeleb2 dataset with total train data of 12.207 utterances, total validation data of 3.618 utterances, and total testing data of 2.068 utterances from 552 speakers. Research done in this paper shows ResNet50 model with x-vector achieved accuracy, recall, precision, and f-1 score of 74.12%, 73.42%, 78.02%, and 0.72 respectively, whereas ResNet50 model with d-vector achieved accuracy, recall, precision, and f-1 score of 36.17%, 35.77%, 36.86%, and 0.32 respectively, whereas the current state-of-the-art model achieved accuracy, recall, precision, and f-1 score of 10.97%, 10.61%, 6.64%, and 0.07 respectively.
Keywords:	Speaker Recognition, Text Independent Speaker Recognition, ResNet50 Model, X-Vector
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	BAG OF VISUAL WORDS AND CNN APPROACHES FOR CONTENT-BASED IMAGE RETRIEVAL USING HOG, GCH AND ORB FEATURES
Author:	K. BHARATHI, DR M CHANDRA MOHAN
Abstract:	Component measurements in digital photos are expanding, and identifying a specific image based on substance from a large database might be difficult at times. A content-based image retrieval (CBIR) method is suggested in this work to extract a feature vector from an image and successfully retrieve content-based pictures. This paper considers three types of image feature descriptor extraction methods: Histogram Oriented Gradients (HOG), Global Colour Histogram (GCH), and Oriented Fast and Rotated BRIEF (ORB). The image feature vectors are kept in the picture database and matched with the testing data feature vector for CBIR at the time of retrieval. In this paper, we aim to present the feature selection based on HOG, GCH and ORB methods to extract features perfectly in capturing the standard dataset CIFAR10 features. The suggested work's execution is evaluated using a Bag of Visual Words, and CNN classifiers. The proposed strategy experiments for different labels indexed elastic search procedures and all cases showed good accuracy in retrieving the correct image.
Keywords:	Information Retrieval, Query Image, Bag Of Visual Words, Elastic Search Engine, Feature Descriptors
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	INVESTIGATING THE ADOPTION OF AN INNOVATION USING AN EXTENDED UTAUT MODEL: THE CASE OF MOBILE LEARNING TECHNOLOGY
Author:	KHALED M. S. FAQIH
Abstract:	Covid-19 pandemic is unleashing unprecedented digital revolution in educational systems, making mobile-enabled learning is unavoidable alternative option, especially in developing countries whereby university students have no option but to use smartphones for learning as desktop and laptop computers are becoming less popular at homes. To achieve the objectives of this study, this study extends the UTAUT model by incorporating three different mechanisms to enhance the UTAUT environment to investigate mobile learning adoption and enrich both theory and practice. The UTAUT theory was extended by incorporating three exogenous variables (perceived compatibility, perceived image, and perceived mobile anxiety), the endogenous variable of perceived innovativeness, and service quality as a moderator. The empirical data was collected using a survey questionnaire administered to higher education students in Jordan. The proposed research model was tested with the use of WarpPLS using 202 useable questionnaires. The results demonstrate that all hypotheses were found statistically significant, indicating that all variables included in this study play an important role in affecting the adoption process of mobile learning. The findings reveal that the research model explains 53% of the variance in the intention to adopt mobile learning. Theoretical contributions and practical implications are discussed.
Keywords:	Higher Education, Mobile Learning, UTAUT, Service Quality, Mobile Anxiety, Jordan
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	A POWER AWARE PROBABILISTIC CLUSTER HEAD SELECTION STRATEGY FOR IOT ENABLED DEVICES ON AGICUTURAL APPLICATIONS FOR TYPESETTING
Author:	PUPPALA TIRUPATHI, NIRANJAN POLALA
Abstract:	IoT enabled devices are becoming highly popular in various domains of industrial and domestic usages. The gain in the popularity is primarily due to the higher adoptability to the situations, effective mobility, availability of the applications for managing data collected from the IoT devices. Also, few of the IoT implementations have demonstrated significant improvements for open IoT stack deployment using Wireless Sensor Networks, which is again accelerated the growth of adaptations. Nonetheless, the IoT enabled device implementations comes with fundamental challenges due to group or cluster-based aggregations. The primary challenge is low battery life due to the higher computational loads and due to the proximity-based load distribution during cluster head-oriented computations. This issue is more prominent for the IoT enabled device deployments for agricultural purposes due to the huge size of the deployment site. Number of research attempts aimed to solve this problem in the recent times. Nevertheless, the existing solutions are highly criticized for higher complexity for deployment and adaptations. Hence, this work proposes a novel solution to balanced workload distribution using probabilistic and regression-based analysis, which results into a highly energy efficient and time efficient cluster head node selection strategy. The probabilistic selection strategy is integrated and optimized with the regression driven analysis to ensure highly random selection and at the same time highly effective selection based on various parameters such as mobility, computational capacity, proximity, and memory consistency. As a result, this strategy demonstrates nearly 92% improvements for time efficiency and nearly 94% improvements for energy efficiency compared with the parallel research outcomes.
Keywords:	Agro-IoT; Probability Distribution; Probabilistic Selection; Equivalency Coefficient; Energy Efficient
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	SPACE-TIME TRELLIS CODE’S PERFORMANCE EVALUATION CONSIDERING BOTH FEED FORWARD AND RECURSIVE STRUCTURE ACROSS WIRELESS CHANNELS
Author:	RAJKUMAR MYLSAMY, KARTHIKA JAYAPRAKASH, SURESHKUMAR ARUMUGAM
Abstract:	The emerging new technologies demands for reliable communication of data at high data rates making wireless communication a challenging field. The major limitations in wireless communications are power, bandwidth and fading. Especially in the applications such as video broadcasting and image transmission requires high data rate and bandwidth. In such applications Space Time Trellis Code Modulation (STTCM) is used to improve the data rate, power efficiency, coding gain and diversity gain by maximizing spatial diversity. The available Space Time Trellis code Modulation technique uses MPSK STTC with different states and non-iterative decoder .In this paper, we iteratively decode MPSK STTCM with various states and with various number of receiving antennas. And the performance of non-iterative decoder and iterative decoder is analyzed individually and together. Based on the analysis some conclusion is obtained. The non-iterative decoder (STTCM) uses feed forward encoder structure and Viterbi algorithm. The iterative decoder (ST Turbo TCM) uses recursive encoder structure and Log-Likelihood value. The BER performance of iterative decoder and non-iterative decoder with 4-PSK and 8 PSK modulation is analyzed with various numbers of states and with variety of antennas which receives it. Outcomes of the simulation shows that the performance of the BER can be enhanced by raising the number of receiving antennas and raising the number of states. And the results also shows that the BER performance of Space Time Turbo Trellis codes are better than Space Time Trellis Codes.
Keywords:	Space Time Trellis Code, Feed Forward Structure, Recursive Structure, Space Time Turbo Trellis Code, BER Performance.
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	NSCT BASED PCA AND K-MEANS CLUSTERING BLOCK LEVEL APPROACH FOR SAR IMAGE DE-NOISING
Author:	DAWAR HUSAIN, DR. MUNAUWER ALAM
Abstract:	Visual data are transmitted as the high quality digital images in the major fields of communication in all of the modern applications. These images on receiving after transmission are most of the times corrupted with noise. This thesis focused on the work which works on the received image processing before it is used for particular applications. We applied image denoising which involves the manipulation of the Nonsubsampled Contourlet Transform (NSCT) coefficients of noisy image data to produce a visually high standard denoised image. This works consist of extensive reviews of the various parametric and non parametric existing denoising algorithms based on statistical estimation approach related to wavelet transforms connected processing approach and contains analytical results of denoising under the effect of various noise at different intensities .These different noise models includes additive and multiplicative types distortions in images used. It includes Gaussian noise and speckle noise. The denoising algorithm is application independent and giving a very high speed performance with desired noise less image even in the presence of high level distortion. Hence, it is not required to have prior knowledge about the type of noise present in the image because of the adaptive nature of the proposed denoising algorithm.
Keywords:	Image - Denoising, NSCT, Gaussian Noise , PCA ,K Mean Clustering.
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	IMPROVING SECURITY AND IMPERCEPTIBILITY USING MODIFIED LEAST SIGNIFICANT BIT AND FERNET SYMMETRIC ENCRYPTION
Author:	EZRA KARUNA WIJAYA, RICO KUMALA, BENFANO SOEWITO
Abstract:	In our daily life in this digital era, information has become essential. Along with the development of technology and the importance of information, digital crimes such as theft of information also develop. One way to protect information is with steganography. Steganography is a technique to insert secret data or secret messages into an image. In this paper, we will modify the Least Significant Bit method so that the secret message will only be entered into one color bit and combine our proposed Least Significant Bit method with cryptography to improve the security and quality of the holding image. The cryptography technique that we will use is fernet symmetric encryption. In this study, we use the Python programming language. From the results of our research, the quality of the holding image increases for the better. We can express this based on the average PSNR value for each image which increases by 0.1% and decreases from the average MSE value, 2.1%. By implementing our proposed method we are able to improve security and imperceptibility proved by the RGB histogram which is very similar to the original images RGB histogram.
Keywords:	Steganography, Cryptography, Least Significant Bit, Fernet Symmetric Encryption, PSNR, MSE
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	FORMATION OF INDIVIDUAL TRAJECTORIES OF GIFTEDNESS OF STUDENTS BASED ON THE ANALYSIS OF LARGE DATA ARRAYS
Author:	ASSEL BEKTENOVA, NATALYA DENISSOVA, LEONID BOBROV
Abstract:	This article discusses the issue of identifying giftedness, that is, the formation of a student's individual trajectories. The relevance of the problem lies in the development of a new paradigm of the information model, which allows to determine the degree of giftedness that meets the needs of digital education and the formation of trajectories for the development of giftedness among students. The article describes the methods of applying fuzzy logic to determine the student's individual trajectories. To do this, in this paper, an attempt was made to use the mathematical apparatus of fuzzy logic. The structure of the database covers all levels of skills in terms of the main indicators of the quality of education. The expert study was carried out on the basis of the Nazarbayev Intellectual School of Chemistry and Biology in Ust-Kamenogorsk. The results obtained made it possible to identify the multi-criteria formation of the trajectories of a gifted student.
Keywords:	Giftedness, Methods For Determining Giftedness, Fuzzy Logic Model, Rules, Individual Trajectory
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	AN IMPROVED DEEPFAKE DETECTION METHOD BASED ON CNNS
Author:	DAFENG GONG, YOGAN JAYA KUMAR, ONG SING GOH, CHOO YUN HUOY, ZI YE5, WANLE CHI
Abstract:	Today's image generation technology can generate high-quality face images, and it isnt easy to recognize the authenticity of the generated images through human eyes. This study aims to improve deepfake detection, a face swapping forgery, by absorbing the advantages of deep learning technologies. This study generates a unified and enhanced data set from multiple sources using spatial enhancement technology to solve the problem of poor detection performance on cross-data sets. Taking the advantages of Inception and ResNet networks, new deepfake detection architecture composed of 20 network layers is proposed as the deepfake detection model. To further improve the proposed model, hyperparameter values are optimized. The experiment result shows that the proposed network significantly enhanced over the mainstream methods, such as ResNeXt50, ResNet101, XceptionNet, and VGG19, in terms of accuracy, loss value, AUC, numbers of parameters, and FLOPs. Overall, the methods introduced in this study can help to expand the data set, better detect deepfake contents, and effectively optimize network models.
Keywords:	Face Swapping, Cross Data set, Deepfake Detection, Data Enhancement, Optimized Hyperparameters
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text

Title:	MODIFICATIONS OF AN ENCRYPTED-BASED SQL MODELS FOR MULTILEVEL DATABASE
Author:	AHMED Y. MAHMOUD, MOHAMMED M. ABU-SAQER
Abstract:	Information security is becoming increasingly vital in a variety of applications. Database applications are no exception, and protecting information based on encryption plays a vital role in the protection of sensitive data. Securing data in several disciplines is a critical concern. Encryption was applied as a mechanism to secure data by database management systems and their applications. A promising approach has been seen in adopting multilevel security and encryption based on multilevel security. Recently, there has been an adaptation of the structure query language to be used in the encryption based on multilevel security; the used models are very time-consuming, especially for the large size of the database. In this paper, we proposed modifications of the models of select, update, and delete operations. The proposed modifications enhanced the performance of the models of select, update, and delete operations. The modification relies on the fact that we avoided repeating the decryption process and replaced it with a single encryption process. The obtained results show that the performances of the proposed modifications are better than the original model.
Keywords:	Multilevel Database, Encryption, Decryption, SQL Models, System Protection
Source:	Journal of Theoretical and Applied Information Technology 15^th September 2022 -- Vol. 100. No. 17-- 2022
Full Text