		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
December 2019 | Vol. 97 No.23

Title:	CLASSIFYING ARABIC TEXT USING DEEP LEARNING
Author:	MOHAMED GALAL, MAGDA M. MADBOULY, ADEL EL-ZOGHBY
Abstract:	Nowadays, the volume of data offered on the Internet is growing every moment, and the necessity to analyze these data and convert to useful information increased. There are several types of research exploring techniques to deal with Text Classification (TC) in many languages; however, In Arabic, the researches are limited. TC is the process of categorizing text document into classes or categories according to the text contents. This research will focus on classifying Arabic Text using a Convolution neural network (CNN), which considered one of deep learning (DL) methods, as it achieved an excellent result in different Natural language processing (NLP) project types [1],[2],[3]. We also introduced a novel algorithm to group similar Arabic words based on extra Arabic letters and word embeddings distances. We named this algorithm as GStem.
Keywords:	Arabic Text Classification, Gstem, Neural Network, Deep Learning
Source:	Journal of Theoretical and Applied Information Technology 15^th December 2019 -- Vol. 97. No. 23 -- 2019
Full Text

Title:	AN INTELLIGENT AGENT FOR MONITORING STUDENTS BEHAVIOR
Author:	HANI HOSNI , MOHAMED A. EL-DOSUKY , MOHAMED EISA ,FIFI FAROUK
Abstract:	This paper surveys Multi Agent Architecture, and then it proposes an agent-based personalized E-learning system. This system is implemented then students are allowed to enroll. The system monitors their behavior and produces statistical reports. The agent of system provides new and important features that are not available in the e-learning systems currently in use.
Keywords:	Multi Agent, Architecture, Behaviour, E-Learning
Source:	Journal of Theoretical and Applied Information Technology 15^th December 2019 -- Vol. 97. No. 23 -- 2019
Full Text

Title:	AN INTELLIGENT AND REAL-TIME RANSOMWARE DETECTION TOOL USING MACHINE LEARNING ALGORITHM
Author:	HIBA ZUHAIR, ALI SELAMAT
Abstract:	Zero-day ransomware still threaten users and enterprises survival in the cyber-space by disturbing electronic amenities, damaging information systems, and causing data and money losses. The publically used anti-ransomware software are trying to mitigate this security issue, however they are limited at identifying zero-day ransomware variants effectively in the real-time without performance overhead. Thus, this paper proposed intelligent, real-time, and three-tier model of ransomware detection tool to be performed well for protecting windows-based information systems. The proposed ransomware detection tool comprises a hybrid machine learning algorithm which hybridizes the decisive functions of two topmost machine learning algorithms (Naïve Bays and Decision Tree) to holistically characterize and accurately classify zero-day ransomware variants in real-time application. Empirical, comparative and realistic assessments demonstrate the adaptability and effectiveness of the proposed ransomware detection tool versus zero-day ransomwares. It achieves approximate accuracy rate of (96. 27%) and mistake rate of (1.32%) along with low misclassifications throughout real-time practice.
Keywords:	Zero-day ransomwares, Signature-based detection, Anomaly-based detection, Hybrid-based detection, Dynamic traits, Hybrid machine learning algorithms.
Source:	Journal of Theoretical and Applied Information Technology 15^th December 2019 -- Vol. 97. No. 23 -- 2019
Full Text

Title:	PARTICLE SWARM OPTIMIZATION AND SIMULATED ANNEALING TO DESIGN BANKRUPTCY PREDICTION NEURAL NETWORK
Author:	FATIMA ZAHRA AZAYITE, SAID ACHCHAB
Abstract:	Bankruptcy prediction models are one of the most interesting subjects in financial engineering research, especially for investors and creditors. In this paper, the Artificial Neural Network is explored as a powerful tool to predict firms’ failure. However, defining the appropriate topology with a suitable set of parameters can be treated as an optimization problem In this study, we investigate the use of Particle Swarm Optimization and Simulated Annealing to develop a performant learning algorithm. The proposed learning algorithm uses an evolved Particle Swarm Optimization algorithm to ameliorate the convergence of the standard algorithm and Simulated Annealing to escape from local minima. Moreover, the leaning algorithm evolves at the same time the number of hidden neurons and the weight values to design the optimum topology. A comparative performance study with Multiple Discriminant Analysis as well as Classification and Regression Tree is reported. The results showed that the proposed model performs better in predicting firms’ Bankruptcy.
Keywords:	Particle Swarm Optimization, Simulated Annealing, Artificial Neural Networks, Machine Learning, Bankruptcy
Source:	Journal of Theoretical and Applied Information Technology 15^th December 2019 -- Vol. 97. No. 23 -- 2019
Full Text

Title:	AN ENHANCED EXTRACTIVE TEXT SUMMARIZATION METHOD FOR MULTIPLE DOCUMENTS
Author:	ADIBA MAHJABIN NITU, MD. PALASH UDDIN, PRIYANKA BASAK TUMPA, SABINA YEASMIN, MASUD IBN AFJAL
Abstract:	Nowadays, text summarization has become an important issue to extract the required information within short time. Several techniques on extractive text summarization have been developed for summarizing English text(s). However, there is a few works done for the summarization of Bengali text(s). In this paper, an improved extractive Bengali text summarization technique has been proposed with enhancing the word scoring process, position value heuristics and summary generation procedure of our previously presented summarizer. In the word scoring procedure, each word is preprocessed using noise removal, tokenization, stop word removal and stemming operation. Then, a heuristics is applied to calculate the word score through checking it in all the input document(s). Moreover, a modified heuristic is proposed for the sentence scoring in which it has given the priority highest to the middle sentence and then the upper and lower sentences from the middle sentence will be less prioritized. Finally, top k-sentences are extracted from each of the clusters of sentences made by K-means clustering algorithm and then the extracted sentences are sorted as their actual appearances in the original document(s). Thus, the final summary is synchronized with the original document(s). In comparison to the existing method, the experimental result shows that the proposed improved technique produces better summarization to satisfy the end-users.
Keywords:	Text Summarization, Extractive Summarization, Bengali Text Summarization, Heuristics, Synchronized Summary
Source:	Journal of Theoretical and Applied Information Technology 15^th December 2019 -- Vol. 97. No. 23 -- 2019
Full Text

Title:	A DIMENSIONAL REDUCED MODEL FOR THE CLASSIFICATION OF RNA-SEQ ANOPHELES GAMBIAE DATA
Author:	MICHEAL OLAOLU AROWOLO, MARION ADEBIYI, AYODELE ADEBIYI
Abstract:	A significant application of gene expression RNA-Seq data is the classification and prediction of biological models. An essential component of data analysis is dimension reduction. This study presents a comparison study on a reduced data using Principal Component Analysis (PCA) feature extraction dimension reduction technique, and evaluates the relative performance of classification procedures of Support Vector Machine (SVM) kernel classification techniques, namely SVM-Polynomial kernels and SVM-Gaussian kernels. An accuracy and computational performance metrics of the processes were carried out. A malaria vector dataset for Ribonucleic Acid Sequencing (RNA-Seq) classification was used in the study, and 99.68% accuracy was achieved in the classification output result.
Keywords:	RNA-Seq, PCA, SVM-Gaussian Kernel, SVM-Polynomial Kernel, Malaria Vector
Source:	Journal of Theoretical and Applied Information Technology 15^th December 2019 -- Vol. 97. No. 23 -- 2019
Full Text

Title:	AUTOMATIC SEMANTIC SENTIMENT ANALYSIS ON TWITTER TWEETS USING MACHINE LEARNING: A COMPARATIVE STUDY
Author:	SARAH M. ALSUBAIE , KHOLOUD M. ALMUTAIRI , NAJLA A. ALNUAIM, REEM A. ALMUQBIL , NIDA ASLAM, IRFAN ULLAH
Abstract:	Due to multiple reasons, social media and microblogs have gained a lot of interest from researchers in the field of Sentiment Analysis recently. Social media platforms comprise one of the most perfect environments of speech and mind expression. This study aims to perform Sentiment Analysis on Twitter platform to identify the polarity of tweets involved in a trending hashtag or event in Twitter. The chosen method for this study is to use ensemble Machine Learning approach using Naïve Bayesian combined with Support Vector Machine, followed by semantic analysis to improve its accuracy. The outcome of the proposed model will be able to determine the polarity of any given text "tweet" to generate a comprehensive statistical report regarding the public's opinion in a certain matter. These reports can be beneficial to marketing specialists, managers, and even Governments to collect the population thinking in order to enhance the standards of living in a region.
Keywords:	Sentiment Analysis (SA), Twitter, Multinomial Naïve Bayes (Multinomial NB), Support Vector Machine (SVM), Classifier Ensemble (CE), Machine Learning (ML)
Source:	Journal of Theoretical and Applied Information Technology 15^th December 2019 -- Vol. 97. No. 23 -- 2019
Full Text

Title:	ENERGY EFFICIENCY OF MULTI-LTE MACRO CELL CELLULAR NETWORKS: MODELLING AND ANALYSIS
Author:	MOHAMMED H. ALSHARIF
Abstract:	This paper evaluates the impact of multi-macro cell systems on the energy efficiency of Long Term Evolution (LTE) cellular networks. Both the proposed model and the analysis of the EE in this study take into account (i) the path losses, fading, and shadowing that affect the received signal at the UE within the same cell, and (ii) the interference effects of adjacent cells. The simulation results show that the interference from adjacent cells can degrade the EE of a multi-cell cellular network. With the high interference from cell2 and cell3 (at the edge of the cell1), the number of bits that will be transferred per joule of energy is 0.78 kb/J with a 1.4 MHz bandwidth and two transmit antennas. With a 20 MHz bandwidth and two transmit antennas, the transfer rate increases to 11.17 kb/J. However, the EE will improve if the number of antennas is increased. The results of this study provide insight into the impact of the number of antennas and the interference from adjacent cells on achieving real gains in the EE of multi-cell LTE cellular networks.
Keywords:	Small cell; Macro-Cell LTE; Data rate; Energy efficiency; Green radio; ICT
Source:	Journal of Theoretical and Applied Information Technology 15^th December 2019 -- Vol. 97. No. 23 -- 2019
Full Text

Title:	A DATA CONFIDENTIALITY SYSTEM BASED ON TRUSTED PLATFORM MODULE IN CLOUD STORAGE ENVIRONMENT
Author:	IGARRAMEN ZAKARIA, HEDABOU MUSTAPHA, BENTAJER AHMED
Abstract:	Data confidentiality is a major concern in cloud storage environment security. A number of methodologies and algorithms are available to prevent privacy vulnerabilities and achieve data security. Existing solutions to protect the data mainly rely on cryptographic techniques. However, these cryptographic techniques add computational overhead, in particular when the data is distributed among multiple Cloud Service Provider (CSP) servers and more precisely Key Management Servers (KMS). File Assured Deletion (FADE) is a promising solution for addressing this issue. FADE achieves assured deletion of files by making them unrecoverable to anybody, including those who manage the cloud storage. The system is built by encrypting all data files before outsourcing, and then using a trusted party to outsource the cryptographic keys. But, this methodology remains weak since its security relies completely on the security of the key manager. In this paper, we propose a new scheme that aims to improve the security of FADE by using the Trusted Platform Module (TPM) and the Encrypted File System (EFS). A prototype implementation of the proposed scheme shows unique results, it provides a value-added security layer compared to FADE with a less overhead computational time.
Keywords:	Cloud Computing, FADE, TPM, VANISH, SSP, Ephemeriser. Cloud Storage, Secure Deletion, Confidentiality, Reliability, Integrity, Trusted Storage.
Source:	Journal of Theoretical and Applied Information Technology 15^th December 2019 -- Vol. 97. No. 23 -- 2019
Full Text

Title:	NEW STABILITY FORMULAS FOR GEOTEXTILE TUBES AGAINST WAVE ATTACK
Author:	YASSINE AMALLAS , YOUSSEF AJDOR
Abstract:	Geotextile tubes are more and more used in maritime field as protection systems (breakwaters, groin, and artificial reef…). They are constantly subjected to ocean hydrodynamic forces. Studying stability of these structures against these agents is a very important task in their dimensioning process because a lack of stability will destroy theme. In literature, there are not yet enough safe, certain and approved formulae or methods to studying stability against wave attack. This is the reason why designers around the world still prefer work with traditional solution (conventional breakwaters with rocks) than these innovative ones. This paper proposes two new formulas to studying stability of geotextile tubes against wave attack. A new called number of stability depending on the degree of filling, the average period and the duration of storm is carried out.
Keywords:	Geotextile Tubes, Stability Formulas, Wave Attack, Number Of Stability.
Source:	Journal of Theoretical and Applied Information Technology 15^th December 2019 -- Vol. 97. No. 23 -- 2019
Full Text

Title:	A MULTI-LAYER SYSTEM FOR SEMANTIC RELATEDNESS EVALUATION
Author:	WAEL HASSAN GOMAA
Abstract:	Measuring semantic relatedness between sentences has always been a major point of discussion for NLP researchers. Semantic relatedness measures are key factors in text intelligence applications as paraphrase detection, short answer grading and information retrieval. This work highlights the effect of investing multiple similarity features by presenting a hybrid multi-layer system where each layer outputs a different independent similarity feature that are then merged using a simple machine learning model to predict text relatedness score. The system layers cover string-oriented, corpus-oriented, knowledge-oriented and sentences embeddings similarity measures. The proposed model has been tested on Sick data set that contains 9840 English sentence pairs. Experiments confirmed that using multiple similarity features is significantly better than applying each measure separately.
Keywords:	Semantic Relatedness, Sentence Embeddings, Text Similarity, Skip-Thought Vector, InferSent
Source:	Journal of Theoretical and Applied Information Technology 15^th December 2019 -- Vol. 97. No. 23 -- 2019
Full Text

Title:	ENHANCED VEHICLE DETECTION APPROACH USING DEEP CONVOLUTIONAL NEURAL NETWORKS
Author:	HOANH NGUYEN
Abstract:	Vehicle detection plays an important role in autonomous driving systems. Recently, vehicle detection methods achieved large successes with the fast development of deep convolutional neural network (CNN). However, due to small size, heavy occlusion or truncation of vehicle in an image, recent CNN detectors still show a limited performance. This paper presents an improved framework based on deep CNN for vehicle detection. Firstly, deconvolutional modules are added at multiple output layers of the reduced VGG 16 architecture to enhance additional context information which is helpful to improve the detection accuracy. Secondly, region proposal modules are applied at different feature maps to address the vehicle occlusion challenge. Due to heavy object occlusion in test dataset, soft Non-Maximum Suppression (NMS) algorithm is used to solve the issue of duplicate proposals. Finally, a deep CNN-based classifier including a region of interest (ROI) pooling layer and a fully connected (FC) layer is used for classification and bounding box regression. The proposed method is evaluated on the KITTI vehicle dataset. Experimental results show that the proposed method achieves better performance compared to other the state-of-the-art approaches in vehicle detection.
Keywords:	Vehicle Detection, Convolutional Neural Network, Intelligent Transportation Systems, Object Detection, Deep Learning
Source:	Journal of Theoretical and Applied Information Technology 15^th December 2019 -- Vol. 97. No. 23 -- 2019
Full Text

Title:	MODIFIED SINE WAVE BASED MODEL IN MADURESE BATIK PATTERN GENERATION
Author:	PURBA DARU KUSUMA
Abstract:	Madurese batik is one of most popular batik besides Solo and Pekalongan because of its richness in color and its uniqueness in adopting plant morphology as its object. Unfortunately, many craftspersons who produce Madurese batik are low income women. It is because of two reasons: most of Madurese batik is handmade batik which is different from Pekalongan batik which is wellknown for machine based printed batik. This condition makes improvization in Madurese batik pattern tends to low. Based of this problem, in this work, we develop batik pattern model that identical with Madurese batik pattern. This model is developed by using sinusoid or sine wave as its basis model rather than popular fractal model. In this work, we focus on developing plant based objects, such as: twig, leaf, and flower. In this work, we also use mostly deterministic approach rather than stochastic approach as in our previous works for the modeling process. Meanwhile, in the batik pattern application, we add some stochastic aspects to control the appearance of some objects. The proposed models then are implemented into web based Madurese batik pattern generation application. Based on the result analysis, sine wave based batik pattern model performs better in creating curvier, smoother and more predictable pattern.
Keywords:	Madurese batik, Sinusoid, Plant morphology, Deterministic.
Source:	Journal of Theoretical and Applied Information Technology 15^th December 2019 -- Vol. 97. No. 23 -- 2019
Full Text

Title:	HANDWRITTEN BENGALI CHARACTER RECOGNITION THROUGH GEOMETRY BASED FEATURE EXTRACTION
Author:	MOSHIUR RAHMAN, IQBAL MAHMUD, MD. PALASH UDDIN, MASUD IBN AFJAL, MD. AHSAN HABIB, FAISAL KABIR
Abstract:	Unlike English characters, one of the major drawbacks in recognizing handwritten Bengali script is the massive amount of characters in Bengali language and their complex shapes. There are 50 complex shaped characters in Bengali alphabet set and working with this huge amount of characters with an appropriate set of feature is a tough problem to solve. Moreover, the ambiguity and precision error are common in handwritten words. In addition, among the huge amount of complex shaped characters, some are very similar in shape those possess severe difficulty to recognize handwritten Bengali characters. Bearing in mind the complexity of the problem, an efficient approach for recognizing handwritten Bengali alphabet is proposed in this work. This proposed approach for identifying Bengali characters is based on character geometry-oriented feature extraction for different handwritten characters. In this paper, different image processing steps are used including image acquisition, digitization , preprocessing, segmentation and feature extraction for tackling the difficulty. Most importantly, the geometry based feature extraction method has been employed to extract the effective features from the Bengali characters for the classification purposes. Then, the classification result was measured for SVM and Artificial Neural Network (ANN) based classifiers on self-generated training and testing data sets which contain 2500 different samples of 50 characters in the Bengali character-set. The proposed technique produces an average recognition rate of 84.56% using SVM and 74.47% using ANN.
Keywords:	Bengali alphabets, image segmentation, feature extraction, support vector machine, artificial neural network
Source:	Journal of Theoretical and Applied Information Technology 15^th December 2019 -- Vol. 97. No. 23 -- 2019
Full Text

Title:	DESIGN OF A CONTEXT-AWARE RECOMMENDER SYSTEMS FOR UNDERGRADUATE PROGRAM RECOMMENDATIONS
Author:	V.VAIDHEHI , R.SUCHITHRA
Abstract:	There are a variety of undergraduate programs available in the education system. Therefore, designing a recommender system based on academics performance alone may not be always helpful. Thus, there is a high demand for knowing other contextual information of the user which can influence the efficiency of the recommender system. Intelligent systems can be designed to predict the contextual information about the user. Selecting the most appropriate undergraduate program by considering different contextual parameters is highly needed for students who have passed class 12. This research work is to design a context-aware recommender system, which recommends under-graduate programs to students of class 12 based on the academic performance and with contextual parameters like financial background, Knowledge level, Group, interested-subject and interested-profession using collaborative filtering approach. This research paper proposes a novel method for creating a rating matrix and the identification and processing of contextual information in an efficient manner. This Context-aware Recommender system is designed based on the predictive values for the various contextual parameters using a contextual modeling approach. Implicit ratings are calculated using the collaborative approach. The results indicate that context-aware recommender engine is more efficient in generating the recommendations thereby improving the user satisfaction level.
Keywords:	Recommender System, Contextual Parameters, Context-Aware Recommender System, Contextual Modeling, Rating Matrix
Source:	Journal of Theoretical and Applied Information Technology 15^th December 2019 -- Vol. 97. No. 23 -- 2019
Full Text

Title:	ESTIMATION OF THE SUCCESS OF BITCOIN AS CRYPTOCURRENCY
Author:	JESÚS ÁLVAREZ CEDILLO, TEODORO ÁLVAREZ SÁNCHEZ , MARIO AGUILAR FERNANDEZ, RAUL JR. SANDOVAL GOMEZ , ANDRÉS CALVILLO TÉLLEZ
Abstract:	Bitcoin is a currency and a digital system. As currency serves for everything that any currency serves, without belonging to a government entity that issues and supports it, is based entirely on the digital system that was designed by its creator, Satoshi Nakamoto. The Bitcoin does not belong to any country or government; and since its creator is anonymous and decided that his invention was a free license, it does not belong to any individual or private company. Also, those who keep their platform running are the users themselves. Investing in bitcoins is precisely the same as making investments with other currencies; the principles are basic: buy cheap and sell expensive and with it is possible to obtain again, but in the case of digital currency there are important considerations because it is a virtual currency. The first of these is its volatility. The price of a bitcoin fluctuates several times during the day in greater magnitude than other currencies or stocks, which is a risk for most, but an opportunity for speculators who know their markets. In this work, an estimation of the success of this cryptocurrency was made in the next ten years. A complete mathematical analysis is shown based on time series.
Keywords:	Bitcoin, Financial System, Micropayments, Cryptocurrencies, Economic Forecasts
Source:	Journal of Theoretical and Applied Information Technology 15^th December 2019 -- Vol. 97. No. 23 -- 2019
Full Text

Title:	MODELING AND QUERYING SPATIOTEMPORAL MULTIDIMENSIONAL DATA ON SEMANTIC WEB: A SURVEY
Author:	IRYA WISNUBHADRA, SAFIZA SUHANA KAMAL BAHARIN, NANNA SURYANA HERMAN
Abstract:	The usage of “web of data” for decision making has increased with the presence of On-Line Analytical Processing (OLAP), Data Warehouse (DW), Multidimensional Data (MD), and Semantic Web (SW) technologies. These technologies are converging into technology that utilizes data on the web to obtain important information as the basis of crucial decision making. The implementation of these technologies continues to grow along with data published on the web using vocabularies like SDMX, QB, and QB4OLAP for linked cube data. Along with increasing analysis complexity, spatiotemporal OLAP emerges as a tool to obtain sophisticated, better, and more intuitive analysis results than OLAP. Vocabulary for spatial OLAP on the Semantic Web has been constructed, namely QB4SOLAP, and successfully implemented. Query language extension for SW was built significantly, but the fundamental model of more dynamic spatial (spatiotemporal) multidimensional data for OLAP on the SW still lacks to exhibit and implemented, even Spatiotemporal DW has been widely studied. This paper presents state-of-the-art research results and outlines future research challenges in Spatiotemporal multidimensional data on the semantic web. This paper organized into three parts, the first part (1) discusses the convergence of OLAP / DW and SW, the second part (2) discusses DW, and spatiotemporal DW on the SW based on the model and the query, and (3) discusses future research opportunities.
Keywords:	Spatiotemporal, Multidimensional Data, Semantic Web, Modeling, Querying
Source:	Journal of Theoretical and Applied Information Technology 15^th December 2019 -- Vol. 97. No. 23 -- 2019
Full Text