		FeedBack \| Contact Us \| Links \| Site Map

Welcome to Journal of Theoretical and Applied Information Technology

Submit Paper / Call for Papers

Journal receives papers in continuous flow and we will consider articles from a wide range of Information Technology disciplines encompassing the most basic research to the most innovative technologies. Please submit your papers electronically to our submission system at http://jatit.org/submit_paper.php in an MSWord, Pdf or compatible format so that they may be evaluated for publication in the upcoming issue. This journal uses a blinded review process; please remember to include all your personal identifiable information in the manuscript before submitting it for review, we will edit the necessary information at our side. Submissions to JATIT should be full research / review papers (properly indicated below main title).

Journal of Theoretical and Applied Information Technology
October 2020 | Vol. 98 No.19

Title:	APPLICATION OF HIERARCHICAL TIME SERIES MODEL WITH TRANSFER FUNCTION
Author:	RAHMA FITRIANI , ACHMAD EFENDI , BAYU RAHAYUDI
Abstract:	The prediction of the amount of water needed in each region is important since clean water is currently needed by every community worldwide. The monthly water consumption depends on how many users. The users vary from people, industry, agriculture, etc. Water providers, in this case local governments, need to estimate the needs in the short- and long-term periods. In the long run, water demand prediction is beneficial for policymakers as the population and water users are increasing such that efforts are needed to supply local water needs. In this study, we aim at predicting the need for monthly water consumption based on users or customers in the city of Malang, East Java, Indonesia and in each district within the city of Malang. There are five districts within this city. The water demand data are stratified from districts and city such that the modeling can accommodate hierarchy. Modeling and forecasting are done with ARIMA (Autoregressive and Moving Average) and transfer function models to accommodate one or several variables affecting water consumption. Accuracy results (Symmetric Mean Absolute Percentage Error) indicate an accuracy error of approximately 3 percent. This result is quite satisfactory and can be used to estimate the accuracy of water demand in Malang city and its districts.
Keywords:	Water, Time Series, Hierarchical, Prediction, Transfer Function
Source:	Journal of Theoretical and Applied Information Technology 15^th October 2020 -- Vol. 98. No. 19 -- 2020
Full Text

Title:	SEGMENTATION OF BRAIN TUMOUR MR IMAGES IN SOFT COMPUTING TECHNIQUES
Author:	C LATHA , DR. K PERUMAL
Abstract:	The fitness function in Genetic algorithm (GA) based Fuzzy C-Means (FCM) and the morphological operations are widely used to extract tumour from MR medical image segmentation, but suffer uncertainty and vagueness in diagnosis. This paper, concentrates on the foremost and important method of segmentation. It is simple and produces a complete division of the image, when applied to medical image analysis, due to sensitivity to noise and poor detection of thin or low signal to noise ratio structure. The present approach helps to correct some drawbacks, on the initial stage of genetic algorithm and probability-based Fuzzy c-means which are close to the original brain images.
Keywords:	Image, Genetic Algorithm, Segmentation, Brain Image, Probability Based Fuzzy C-Means, Morphological Operations
Source:	Journal of Theoretical and Applied Information Technology 15^th October 2020 -- Vol. 98. No. 19 -- 2020
Full Text

Title:	BAYESIAN APPROACH FOR COMPETENCE FORMATION FOR STUDENTS OF IT-SPECIALTY
Author:	ASEM SHAYAKHMETOVA , PERIZAT SEISENBEKOVA , MOHAMED OTHMAN , ORKEN MAMYRBAYEV , DINARA KASSYMOVA
Abstract:	The Bayesian networks theory has recently become very popular in solving various applied problems in multiple fields of science and industry. For the practical application of the Bayesian approach, a high quality software product that implements the mathematical theory of Bayesian networks is required. The Bayesian approach is a promising approach for creating an intelligent environment to enhance student competence. To implement Bayesian networks, the BayesiaLab application software package is well suited and is one of the high-quality software products, which is specialised in artificial intelligence technologies. With the help of the BayesiaLab package, various models of Bayesian networks can be created, explored, edited and analysed. This article introduces student competences and explores the possibilities of using Bayesian networks in the formation of the competences of information technology (IT) students and for this purpose, a general algorithm and a specific architecture of the intellectual environment have been developed. It is a known fact that improved professional competence in education increases the competitiveness of specialists and updates the corresponding educational environment.
Keywords:	Bayesian Approach, Competence Formation, IT Students, Education, Bayesialab package
Source:	Journal of Theoretical and Applied Information Technology 15^th October 2020 -- Vol. 98. No. 19 -- 2020
Full Text

Title:	REAL-TIME DISSOLVED OXYGEN MONITORING BASED ON THE INTERNET OF SMART FARMING PLATFORM
Author:	WASANA BOONSONG, OLUSEYE ADELEKE, JUTHAMAS JANTHOTHAI, KANLAYA RACHALEK
Abstract:	The Internet of Things (IoT) has not only become a part of human life, it has become a tool being made important, especially in the 21st century for generating revenue and strengthening communities’ economies. Smart farming on the other hand, is a management concept that is aimed at providing the agricultural industry with a system or infrastructure to leverage on advanced technology such as the cloud, IoT and even big data – for tracking, automating and monitoring operations. This study focus is on using the IoT on a smart farming platform for dissolved oxygen monitoring for aquaculture in the Songkhla Lake Basin, South of Thailand. We propose the fundamentals of real-time dissolved oxygen (DO) monitoring based on the internet of smart farming platform, in which the parameters studied are DO and water temperature (TEMP). The system is built to automatically detect data from the aquaculture pond or natural water sources, after which the data detected is sent to the host computer and/or user smart phone through the Wi-Fi cloud service network. The data collected for the DO and TEMP show that the relationship of both parameters are negatively related. In other words, when the water temperature rises, the DO value decreases accordingly and vice versa. The results were analyzed by the Pearson correlation statistical program at 99 percent confidence of -.732**.
Keywords:	Iot, Dissolved Oxygen, Pearson Correlation
Source:	Journal of Theoretical and Applied Information Technology 15^th October 2020 -- Vol. 98. No. 19 -- 2020
Full Text

Title:	SYMPTOM-BASED DISEASE PREDICTION SYSTEM USING MACHINE LEARNING
Author:	JING YI LEONG, BOOMA P M
Abstract:	Healthcare has been an important industry from then till now, and it is said to be one of the sectors which plays a critical role in preventing the increasing number of a particular disease. In this era of new technology, machine learning has been used in a lot of industries, and one would be the healthcare industry. In the healthcare field, machine learning contributes significantly to predicting a disease as to simplify the process of the manual disease diagnosis and bring convenience to both the doctor and patient. In this paper, a disease prediction system will be implemented with the use of supervised learning algorithm to allow patient in identifying disease themselves based on their symptoms. Few supervised learning algorithms are being trained and tested in terms of their accuracy, and the algorithm with the highest accuracy is used for the prediction. The chosen supervised learning algorithms to be tested include Bernoulli Naïve Bayes, Decision Tree, and Support Vector Machine.
Keywords:	Classification, Disease, Healthcare, Machine Learning, Prediction.
Source:	Journal of Theoretical and Applied Information Technology 15^th October 2020 -- Vol. 98. No. 19 -- 2020
Full Text

Title:	OPTIMISATION OF A PUBLIC BUS SCHEDULE USING A NEURAL NETWORK AND K-MEANS
Author:	YASUKI SHIMA , RABIAH ABDUL KADIR , ALI FATHELALEM , RIZA SULAIMAN
Abstract:	Public transport bus services are expected to evolve with the growth of a city. In urban areas in both developed and less developed countries, the demand for public transport is gradually increasing, and it is becoming necessary to increase the capacity of these services. In the case of public buses, operators tend to increase the number of vehicles and to extend bus routes to meet the requirements of passengers. However, waiting times for passengers are still critical, and passengers may spend longer waiting for public transport services. In this research work, we use a neural network and K-means to find the optimal scheduling of bus trips per day, and to minimise both the number of buses per day (i.e. the cost) and the waiting time throughout the day. A GPS-based dataset obtained from a public bus operator in Okinawa, Japan, is used. Three elements of the dataset (the number of buses, number of stops and dwell time per time zone) are extracted to identify the peak and off-peak times in each day's bus service using K-means. These three elements are also used as input to the ANN. As the output of K-means clustering, a moderate dwell time is used as a supervised dataset. A backpropagation neural network algorithm is used to optimise the bus schedule and to allocate vehicles per time zone in a way that minimises the operational cost and maintains reliability for passengers. Our research establishes a framework and methods for optimising operating costs while meeting the passenger demand for reliable services and minimal waiting times.
Keywords:	Public Bus, Bus Scheduling, K-Means, Neural Network.
Source:	Journal of Theoretical and Applied Information Technology 15^th October 2020 -- Vol. 98. No. 19 -- 2020
Full Text

Title:	TOWARD INCREASING AND INVESTIGATING E-TOURISM DATA WAREHOUSE THROUGH A WEBSITES ANALYSIS STRATEGY
Author:	S. BOUREKKADI , K. SLIMANI , O. EL IMRANI , M.EZZAKI , A.BABOUNIA , Y.FAKHRI
Abstract:	Nowadays, the competition between companies of all fields focuses on the information provided to the customers. International economic exchanges have experienced battles between large global companies because of the good marketing of their information. This guides us to confirm that the companies that perfectly manage their information related to their activities and know how to take advantage of the data that exist in their field are the companies that can keep an economic continuity. For this reason, the objective of this study is to work on an analysis strategy of websites especially for companies working in the field of tourism. This analysis will enable targeted companies to compete with their peers for informational competition to secure increased and investigated warehouses’ data. This research is based on a multi-axis analysis as well as on a statistical study that collects the opinions of visitors on the used websites as case studies for this research. The main obtained results include the fact that several websites have many weaknesses. This is something that influences their performance. In addition, it was noticed that several websites’ managers think that the performance of their product is based only on the number of visitors per day or the total of pages read in a specific period. In fact, these characteristics add nothing to the analysis when there are micro-conversions leading to macro-conversions.
Keywords:	Five Analysis, E-Tourism, Data Warehouse, Standby Strategy.
Source:	Journal of Theoretical and Applied Information Technology 15^th October 2020 -- Vol. 98. No. 19 -- 2020
Full Text

Title:	TREE VOLUME MODELLLING AND VALIDATION USING MACINE LEARNING APPROACHES FOR FOREST RESERVE
Author:	SITI HAJAR MOHD MUSHAR, SHARIFAH SAKINAH SYED AHMAD, FAUZIAH KASMIN, NUR HAJAR ZAMAH SHARI
Abstract:	Forest plays an important role in national growth as the forestry and logging activities contribute 5.6% to the Malaysia GDP of the agricultural sector in 2018. A precise value of tree volume estimation highly affects forest management and administration. The forest management and administration framework are designed based on the evaluation of the forest, including its current volume; therefore this strongly supports the need for a precise tree volume estimation. Tree volume can be expressed either in terms of the total cubic volume of a tree or in terms of the total cubic volume of an area. However, this paper is going to focus on the volume estimation technique for an individual tree. Analysis of the literature found that the commonly used method in estimating the tree volume is regression, however, growth in the information technology has driven the use of machine learning techniques. The state-of-the-art highlighted that machine learning not only has a high capability in developing a robust model but also able to overcome the regression analysis problem such as overfitting of the data. Numerous comparison studies on the application of machine learning in forest modelling can be found but there are discrepancies of analysis among scholars. Therefore, this paper will perform tree volume estimation by using regression and four machine learning techniques which are artificial neural network (ANN), epsilon-Support Vector Regression (ε-SVR), k-Nearest Neighbor (k-NN) and random forest (RF). The precision and accuracy of the volume model will be verified by using the root mean square error (RMSE) and standard deviation (SD). The result and analysis of this study seem to be consistent with other research which found that machine learning techniques perform better than regression as the ANN is the best modelling technique for dipterocarp and non-dipterocarp datasets while for all species dataset, ε-SVR records the highest accuracy.
Keywords:	Machine Learning, Regression, Volume Model, Tree Volume
Source:	Journal of Theoretical and Applied Information Technology 15^th October 2020 -- Vol. 98. No. 19 -- 2020
Full Text

Title:	GENDER IDENTIFICATION AND AGE ESTIMATION OF ARABIC SPEAKER USING MACHINE LEARNING
Author:	FIRAS IBRAHIM, KHALID M.O. NAHAR, MOY'AWIAH A. AL-SHANNAQ
Abstract:	Gender Identification and Age estimation is an important topic in the field of Automatic Speech Recognition (ASR) systems. In the field of robotics, for example, it is important to identify human sex and age for emotion recognition and robot interaction. In this paper, we targeted Arabic speakers by identifying their genders and estimating their ages. Experiments were conducted using six famous learning algorithms such as NB-Tree (Decision Tree), Random Forest (RF), K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Artificial Neural Network (ANN), and Naïve Bayes (NB). We focused on the accuracy with some important classification measurements. Automatic gender identification and age estimation system is proposed based on extracting MFCC features from Arabic speech. The MFCC features with the machine learning algorithms were applied to determine whether the speech sample is male or female and assign the approximate age slice. Two experiments were done, the first one targeted gender identification. The results of the first experiment showed that the learning algorithms SVM and ANN were superior in gender determination with accuracies 98.5% and 96.5% respectively. The second experiment targeted age estimation. The results of this experiment showed that Decision Tree (NB-Tree), and Random Forest (RF) were superior in age estimation with accuracies 95.9%, and 93.0% respectively.
Keywords:	Age Estimation, Automatic Speech Recognition, Gender Identification, Machine Learning, MFCC, Artificial Neural Network (ANN).
Source:	Journal of Theoretical and Applied Information Technology 15^th October 2020 -- Vol. 98. No. 19 -- 2020
Full Text

Title:	CLASSIFICATION AND CONSTRUCTION OF ARABIC CORPUS: FIGURATIVE AND LITERAL
Author:	NOUH SABRI ELMITWALLY , AHMED ALSAYAT
Abstract:	Annotation of Arabic texts is an imperative task that is also costly and time-consuming. To overcome these obstacles to creating Arabic resources for analysis and training, we have built an integrated Arabic corpus. Constructing an Arabic corpus including various rhetorical and unusual sentences is a challenging task for the classification methods. Such a wide, reliable Arabic corpora doesn’t exist, which has motivated us to create a corpus for further analysis. The main contributions of this paper are twofold: (1) We construct the Arabic Figurative Sentiment Analysis (AFSA) corpus, consisting of the annotated figurative sentiment texts for the Arabic Saudi Dialect and Modern Standard Arabic (MSA). The construction of this corpus is based on the Arabic Language Sentiment Analysis (ALSA) framework, which involves annotated literal and figurative texts. The collected data contains 2000 texts from the Holy Quran, Al-Hadeeth, and the Arabic Saudi Dialect Dataset, which is comprised of 1000 literal and 1000 figurative annotations. (2) The process developed classifies the collected texts into figurative categories resulting in F1-scores reaching 92%, 93%, and 94% using the Multinomial Naïve Bayes (MNB) classifier, logistic regression (LR), and the Bernoulli Naïve Bayes (BNB) approach, respectively.
Keywords:	Machine Learning; Arabic Sentiment Analysis; Figurative Language
Source:	Journal of Theoretical and Applied Information Technology 15^th October 2020 -- Vol. 98. No. 19 -- 2020
Full Text

Title:	A COMPREHENSIVE STUDY OF BOTNETS ON INTERNET OF THINGS AND MOBILE DEVICES : DETECTION AND MITIGATION TECHNIQUES
Author:	SHWETARANI , NAWAB MUHAMMAD FASEEH QURESHI , DONG RYEOL SHIN
Abstract:	Network formed by a group of internet-connected compromised devices is known as Botnet, which may include personal computers, servers, internet of things and mobile devices. The botnet has been one of the most common network security threat. Botnets have been used for stealing data, sending spam, and allows attackers to access the device for collecting personal information of the users and for conducting distributed denial of service attacks (DDoS attacks). Increasing popularity and recent advances in the Internet of Things (IoT) and mobile devices have made IoT devices and mobile devices an easy and alluring target for attackers. Various studies have proposed many sophisticated mechanisms for understanding and identifying botnets and how they creating security threats for IoT devices and Mobile devices. This survey work presents a comprehensive review that discusses about IoT and mobile botnets propagation, detection, and mitigation. In this work, we focus on various types of IoT and mobile botnets propagation, attack methodology, and how they exploited in DDoS attacks along with various technologies used to detect IoT and Mobile botnets. Also we introduce the structure and characteristics of basic botnets.
Keywords:	Botnets, Internet of Things (IoT), IoT botnets, Mobile botnets, DDoS attacks, cybersecurity
Source:	Journal of Theoretical and Applied Information Technology 15^th October 2020 -- Vol. 98. No. 19 -- 2020
Full Text

Title:	FAST AND ACCURATE INSTANCE SEGMENTATION FOR AUTONOMOUS DRIVING BASED ON REGION-BASED CONVOLUTIONAL NEURAL NETWORK
Author:	HOANH NGUYEN
Abstract:	This paper presents a two-stage framework for fast and accurate instance segmentation of objects in traffic scene images based on region-based convolutional neural network. For improving the inference speed of the proposed framework, a lightweight deep convolutional neural network which achieved high accuracy in very limited computational budgets is adopted to generate base feature maps. To enhance the segmentation performance on small objects, this paper designs an enhanced module to generate fused feature map which improves the resolution of small objects and simultaneously includes more semantic information. The fused feature map enhances the classification performance and the segmentation performance of small objects. Furthermore, an improved RoI pooling process based on deformable RoI pooling is proposed in this paper. The improved RoI pooling employs a lightweight offset prediction branch which contains fewer parameters compared with standard offset prediction branch, thus improving the inference speed of the proposed framework. For evaluating the proposed framework on instance segmentation of objects in traffic scene images, the Cityscapes dataset is adopted. Experimental results show the effectiveness of the proposed method on both accuracy and inference speed.
Keywords:	Instance Segmentation, Autonomous Driving, RoI Pooling, Deep Convolutional Neural Network, Region-based Convolutional Neural Network
Source:	Journal of Theoretical and Applied Information Technology 15^th October 2020 -- Vol. 98. No. 19 -- 2020
Full Text

Title:	KEY SUCCESS FACTORS OF BUSINESS PERFORMANCE: EVIDENCE FROM SMART FACTORIES IN SOUTH KOREA
Author:	ROK LEE, SUNG HYEON PARK, JU GYEONG PARK
Abstract:	The purpose of this study is to reveal the effects of key smart-factory success factors—the Internet of Things (IoT), big data, artificial intelligence (AI), fifth-generation (5G), and MES (manufacturing execution systems)—on the continuous use and business performance of smart factories in small and medium-sized manufacturing companies. To achieve this, an empirical survey was conducted with 189 key managers in 50 small and medium-sized enterprises (SMEs) that have established smart factories with government aid, and the responses were analyzed with a structural equation model using the AMOS program. The research findings show that these success factors were only partially effective—they had a positive effect on the continuous use of smart factories by SMEs, as well as the factories’ productivity and quality performance (sub-factors of business performance), but had no significant effect on financial performance. This means that the effective utilization of these factors can enable high-quality performance by maintaining operational efficiency. Consequently, the integration and utilization of IoT, big data, AI, 5G, and MES based on AI technologies have policy and practical implications, showing that they should be used as key success factors in SMEs.
Keywords:	Smart Factory, Key Success Factors, Business Performance, Korea, Small and Medium Enterprises
Source:	Journal of Theoretical and Applied Information Technology 15^th October 2020 -- Vol. 98. No. 19 -- 2020
Full Text

Title:	OPTIMIZATION OF TOURIST TRANSPORTATION
Author:	АNEL BARLYBEKOVNA SHINYKULOVA, YERSAIYN KURMANBAYULY МAILYBAYEV, DMITRY VIKTOROVICH ISAYKIN, IGOR OLEGOVICH KOSYAKOV, UMIRBEK UMBETULY UMBETOV
Abstract:	Transport is one of the most important components of the material base of any country's economy. Since ancient times, transport has been the engine of progress. The person used any available means to transport people and goods. With the invention of the wheel, and a little later of various types of engines, man began to develop vehicles accordingly: carts, coaches, steamers, locomotives, planes, etc. This made it possible to travel long distances and for various purposes. In this research, we explore the optimization of tourist transportation problem. In our research we use branch and bound method, and dynamic programming techniques to ensure optimization in tourist transportation problem.
Keywords:	Transportation, Optimization, Suicidal Ideation Detection, Machine Learning, Social Media
Source:	Journal of Theoretical and Applied Information Technology 15^th October 2020 -- Vol. 98. No. 19 -- 2020
Full Text

Title:	DESIGN FOR ADDITIVE MANUFACTURING
Author:	H.SALEM , H.ABOUCHADI , K. EL BIKRI
Abstract:	Additive manufacturing (AM) is increasingly used in different fields. At first, it was specific to prototyping and proof of concepts. Nowadays, it is used in many areas. AM allows the fabrication of non-removable assemblies in one go, with two or more different materials. The complexity of the parts is not limited with the tool access or other blocking issues of traditional processes. The only limitation is the imagination of the designer. This brings up a change of paradigm when thinking the design of new parts, or the re-engineering of existing assemblies. To benefit from these advantages, a new design approach must be developed; it should take into account the specificities of the process, and help the designer find optimum solutions. The design methodologies have been developed for a long time, they are mostly thought for a specific life cycle or a specific manufacturing process. Because of the differences of AM technologies, the design thinking of these processes is important in the laboratories using AM. The aim of this paper is to present the traditional methodologies, outline the need for a specific one, and present a new methodology concerning the DFAM (design for additive manufacturing), including the factors influencing the design, and the added value compared to the cited methodologies.
Keywords:	Additive Manufacturing, Design, Methodology, Manufacturing Process, 3D Printing
Source:	Journal of Theoretical and Applied Information Technology 15^th October 2020 -- Vol. 98. No. 19 -- 2020
Full Text

Title:	OPTIMIZATION OF FUZZY C-MEANS CLUSTERING USING PARTICLE SWARM OPTIMIZATION IN BRAIN TUMOR IMAGE SEGMENTATION
Author:	RIDHO MAHESA, ERI PRASETYO WIBOWO
Abstract:	Tumor is an abnormal cell development in the body, one of them in the brain. Before performing examine of brain tumor patients, medical team will first analyze the results of medical imaging to find out the part that is a tumor and not the tumor in the medical image. Analysis of tumor segmentation that is still manually requires a longer time, thus inhibiting the treatment process of patients to enter the next treatment stage and delay the latest information about the health of patients. Therefore, a mechanism is needed to automatically segment brain tumor regions. Until now, a lot of research has been done to segment brain tumor regions automatically. Based on research conducted by previous researchers about the segmentation of brain tumor regions, some researchers still use segmentation algorithms that are sensitive to initial position of the cluster center or determining seed points which often contain either too many regions. There is a possibility of getting unfavorable segmentation result. This research aims to propose method to develop an existing partition-based brain tumor segmentation algorithm by adding stages of optimization algorithm of the image segmentation algorithm. In this research, image segmentation algorithm used is Fuzzy C-Means. For optimize Fuzzy C-Means, used Particle Swarm Optimization. Optimization algorithm run concurrently with segmentation algorithm. The performance measurements used by comparing objective function of original algorithm without optimization with 6 images data. As a result, objective function of Fuzzy C-Means optimized by Particle Swarm Optimization (FCM - PSO) achieve more minimum than original Fuzzy C-Means (FCM) of six images data. It means, objective function of Fuzzy C-Means optimized by Particle Swarm Optimization closer to global minimum and can used to optimize segmentation algorithm.
Keywords:	Optimization, Segmentation, Fuzzy C-Means, Particle Swarm Optimization, Brain Tumor
Source:	Journal of Theoretical and Applied Information Technology 15^th October 2020 -- Vol. 98. No. 19 -- 2020
Full Text

Title:	COMPARISON OF STEEPEST ASCENT HILL CLIMBING ALGORITHM AND BEST FIRST SEARCH ALGORITHM IN DETERMINING THE SHORTEST ROUTE FOR MEDAN TOURISM
Author:	DIAN RACHMAWATI, HANDRIZAL, RIZALI AHMAD BATUBARA
Abstract:	Medan is the capital of North Sumatra and one of the major cities in Indonesia. In this city there are several tourist destinations that can be visited by the people. When visitors want to travel by visiting several destinations, it will be difficult for visitors to determine the shortest route that can be traversed. And to find out the shortest travel route, a system is required to find the shortest route that visitors can travel through. In this study developed a route search system for the nearest tourist trip to Medan. In this system there is a menu to find routes by providing several tourist destinations in Medan. And in support of the shortest route search process on this system used the Steep Ascent Hill-Climbing algorithm and the Best First Search algorithm. The comparison process will then be carried out between the two algorithms, both from running time and the distance weight generated by the algorithm used. After both algorithms are implemented into the system and tested both algorithms, then obtained the complexity of the Steepest Ascent Hill Climbing algorithm is θ (n5), and the running time value is 59.3856 ms. On the contrary, obtained the Best First θ (n3) algorithm, and the running time value is 54.6669 ms. And from the testing of both algorithms, it can be concluded that the search using the Best First Search algorithm produces a better running time value than the Steepest Ascent Hill Climbing algorithm.
Keywords:	Best First Search, Steepest Ascent Hill Climbing, The Shortest Route
Source:	Journal of Theoretical and Applied Information Technology 15^th October 2020 -- Vol. 98. No. 19 -- 2020
Full Text

Title:	A COMPARATIVE ANALYSIS OF THE SEVERAL MATRIX FACTORIZATION PROCESS IN IMAGE RECONSTRUCTION AND A HOMOGENEOUS APPROACH TO THE FOURIER SERIES
Author:	A.N.M. REZAUL KARIM
Abstract:	Over the last few decades, image reconstruction has become an interesting field for the development of computer-based applications. The Decomposition of a matrix or matrix factorization is one of the most important components in many engineering and scientific applications. This technique is used to decompose one matrix into more than one matrix. One can efficiently solve a system of equations based on matrix factorization, and this, in turn, is the foundation of the inverse matrix, which is a major component of several important algorithms. Matrix factorizations are widely applied in situations that involve solving linear systems, numerical linear algebra, rank estimation, image processing, image reconstruction etc. This paper attempts to analyze the techniques of matrix factorization or decomposition techniques used in image reconstruction based on their advantages, disadvantages, limitations and computation complexity. Some techniques are examined and a comparative evaluation of these strategies is presented. This paper also shows the homogeneity between the Fourier series and matrix factorization process in image reconstruction.
Keywords:	Matrix Factorization, Image Reconstruction, Fourier Series, Eigen Decomposition, Cholesky Decomposition, LU Decomposition, SVD.
Source:	Journal of Theoretical and Applied Information Technology 15^th October 2020 -- Vol. 98. No. 19 -- 2020
Full Text