		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
November 2017 | Vol. 95 No.21

Title:	USING AN ANT COLONY OPTIMIZATION ALGORITHM FOR IMAGE EDGE DETECTION AS A THRESHOLD SEGMENTATION FOR OCR SYSTEM
Author:	MARYAM ASGARI, FARSHID PIRAHANSIAH, MOHAMMAD SHAHVERDY, MEHDI FARTASH
Abstract:	Binarization or thresholding is one problem that have to solve in pattern recognition methods and applications. Moreover, it has a very important influence on the sequent steps in computer vision applications such as, Optical Character Recognition (OCR), image segmentation, and tracking objects. Ant colony optimization (ACO) is a population-based metaheuristic which use to solve optimizations problems in diverse fields, such as traffic congestion and control, structural optimization, manufacturing, and genomics are presented. In this work, a combination of ant colony, edge detection, and thresholding methods are combine in order to use in OCR system. The algorithm which is used the DIBCO 2009 in printed and a handwritten image was tested. This method has a compare with Kittler and Illingworth's Minimum Error Thresholding, potential difference, max entropy, Pirahansiahs method and Otsu.
Keywords:	Ant Colony Optimization (ACO), Peak Signal-To-Noise Ratio (PSNR), Single Thresholding, Image Processing, Image Segmentation, Optical Character Recognition, Computer Vision, Deep Learning.
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	AN EFFICIENT HECTIC COMPOSITION WEB SEQUENTIAL BASED PATTERN TREES IN LARGE DATABASES
Author:	A.P.SELVA PRABHU, DR.T.RAVICHANDRAN
Abstract:	Web sequential pattern mining identifies frequent subsequences as patterns from large database. In this paper, a novel framework called Hectic Composition Mining based Approximate Pattern Tree (HCM-APT) to handle different access pattern by linking operation is presented. This framework extends the tree based indexing model, which dynamically adjusts links in the mining process using Composite Pattern Mining. A distinct feature of HCM-APT framework is that it clusters on a very limited and precisely predictable space which runs fast in memory based setting. As a result, the framework HCM-APT scales up to very large database through database segregation extensively minimizing the memory space. For dense base, competent Approximate Pattern Trees are constructed dynamically for obtaining rich properties by significantly reducing the execution time for obtaining rich properties. Finally, the proposed framework applies a scalable mining model for approximate patterns generated through tree using Variable Regression function for improving the scalability in mining large databases. Experimental results on Amazon Commerce reviews dataset show the proposed framework HCM-APT outperform other well-established methods in identifying hectic composition pattern. Experiment is conducted on factors such as execution time for obtaining rich properties, memory space consumption and scalability to mine the sequential patterns effectively.
Keywords:	Web Sequential Pattern Mining, Bidirectional Pattern, Access Patterns, Composite Pattern Mining, Approximate Pattern Trees, Variable Regression Function
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	ACCELERATING THE OUTLIER DETECTION METHODS FOR CATEGORICAL DATA BY USING MATRIX OF ATTRIBUTE VALUE FREQUENCY
Author:	NUR ROKHMAN, SUBANAR, EDI WINARKO
Abstract:	Based on the data, outlier detection methods can be classified into three classes. Those are the methods which work on numerical data, work on categorical data, and work on mixed type data. Most of the outlier detection method works on numerical data. Only few method works on categorical data or work on mixed type data. In this paper, a new method for detecting outlier in categorical data called Weighted Matrix Entropy Value Frequency (WMEVF) has been proposed. This method uses weighting function to improve the precision and uses a matrix of attribute value frequency to reduce the complexity. There are four weighting functions used in the experiments namely: range, variations, deviation standard, and square function. The performance of WMEVF is observed based on the detected outlier of UCI Machine Learning datasets and the time needed to detect the outlier. The experiments show the fact that square function improved the precision and the matrix of attribute value frequency reduced the complexcity from O(mn2) to O(mn).
Keywords:	Outlier Detection, Categorical Data, Weighting Function, Entropy, Attribute Value Frequency.
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	Q-LEARNING WITH ADAPTIVE KANERVA CODING ON PROTEIN DOCKING
Author:	ERZAM MARLISAH, RAZALI YAAKOB, MD NASIR SULAIMAN, MOHD BASYARUDDIN ABDUL RAHMAN
Abstract:	Molecular docking is an important process in pharmaceutical research and drug design. It is used in screening libraries of small molecules or ligands to bind to a target protein changing its original biochemical properties forming new stable complex. In docking ligand to protein, the ligands pose, i.e. position, orientation and torsion angles, is translated, orientated, and the ligands torsion angles are rotated repeatedly to find an ideal site on the protein to bind. In this paper, Q-learning algorithm, a model-free reinforcement learning, with adaptive Kanerva Coding is used as the searching algorithm for protein-ligand docking problem. It evaluates the effectiveness of Q-learning algorithm and the different settings for the parameters of reinforcement learning. A popular docking tool called AutoDock Vina was used to find the ligands goal pose. The effectiveness of the agent is measured by the success of finding the goals. The proposed agent managed to match and finds better pose than AutoDock Vina in medium to large size ligands.
Keywords:	Reinforcement Learning, Q-learning, Protein-Ligand Docking, Machine Learning, Kanerva Coding
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	EFFECTIVENESS OF K-MEANS CLUSTERING TO DISTRIBUTE TRAINING DATA AND TESTING DATA ON K-NEAREST NEIGHBOR CLASSIFICATION
Author:	MUSTAKIM
Abstract:	One of the constraints in classification is how to divide the dataset into two parts, training and testing which can represent every data distribution. The most commonly used technique is K-Fold Cross Validation which divides data into several parts and alternately into training data and testing data. In addition, the commonly used technique is to divide data into percentage form (70% and 30%), also become an option in data mining research. K-Means is a grouping algorithm which able to maximizes the effectiveness of distributing data in classification. The experiments performed using K-Means Clustering against K-Nearest Neighbor (K-NN) which was validated by Confusion Matrix have the highest accuracy of 93.4%, it is higher than the K-Fold Cross Validation data distribution technique for each experiment using data Education Management Information System (EMIS) as well as random data. The concept of distributing data in groups can be a representative to each member and increase the accuracy of classification algorithm, although the experiment only applied 70% of training data and 30% of testing data in each group.
Keywords:	Confusion Matrix, K-Fold Cross Validation, K-Means Clustering, K-Nearest Neighbor
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	A ROBUST DUAL SOURCE LEVEL SET METHOD FOR THREE-DIMENSIONAL ECHOCARDIOGRAPHY IMAGE SEGMENTATION
Author:	NIMA SAHBA, EMAD FATEMIZADEH, HAMID BEHNAM
Abstract:	Echocardiography as a common place device has been widely used for diagnosis. Segmentation of left ventricle just using the standard level set formulation based on the magnitude of intensity cannot restrict the contour evolution completely due to the lack of sharp edges. This paper aims to introduce a robust method for segmentation of three-dimensional echocardiography images based on two boundary maps namely distance map and probabilistic map as stopping criteria of contour evolution which are extracted using two different approaches. The distance map is extracted by applying Free Form Deformation (FFD) method on manually determined landmarks of initial spherical volume to register the landmarks to boundary of left ventricle using B-Spline interpolation. The probabilistic map is independently created as a contour guideline to achieve probable desirable regions. Segmentation of echocardiography volumes is implemented using a proposed energy function based on level set formulation combined by two edge maps which is named dual source level set. Left ventricle segmentation using proposed method illustrates expert-approved performance, providing a reliable tool for clinical practice due to less than 2% ejection fraction error compared to manual tracing by expert. Making use of combined distance and probabilistic map as stopping criterion in level set approach helps to avoid the local minima and excessive contour expansion, while the left ventricle valve is open or there is not clear edge.
Keywords:	Level set, Segmentation, three-dimensional echocardiography, Dual Source.
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	MODELLING OF ASPECTS USING ASPECT-ORIENTED DESIGN LANGUAGE
Author:	SAQIB IQBAL, ABDALLA MANSUR, GARY ALLEN
Abstract:	The Aspect-composition is a vital step in aspect modelling. Aspects are composed with each other and with base constructs through pointcuts defined in the aspects. Design languages address this composition by providing composition techniques and directives. However, most of the contemporary design languages lack support for inter-aspect and inner-aspect compositions. Another problem is resolving aspect interference which arises as a result of a composition. Although some techniques have been proposed to overcome aspect interference at the implementation level, the problem needs attention at the modelling level. The eradication of interference and conflicts related to aspect composition at the modelling stage could ensure better implementation and fewer conflicts. This paper provides a composition strategy equipped with new design notations and diagrams to provide support for aspect compositions, as well as inner-aspect compositions. The paper also provides a technique to prioritize aspect execution at the modelling stage to reduce aspect interference and aspect conflicts.
Keywords:	Aspect-Oriented Programming, Pointcut Modelling, Aspect Composition, Aspect-Oriented Model
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	SUSTAINABLE SOFTWARE DEVELOPMENT LIFE CYCLE PROCESS MODEL BASED ON CAPABILITY MATURITY MODEL INTEGRATION: A STUDY IN MALAYSIA
Author:	KOMEIL RAISIAN, JAMAIAH YAHAYA, AZIZ DERAMAN
Abstract:	Green software engineering is an important view of software engineering process in the current century. Previously, software engineers concerned mainly with development of hardware or software without giving more importance to sustainability. The field of green software and green software engineering is still young. Thus, in modern society, researches efforts are mainly focused on green and sustainable software engineering itself. The development of sustainable software has been identified as one of the key challenges in the field of computational science and software engineering and also there no is clear idea regarding how to accomplish the Green and sustainability in Software Development Life Cycle (SDLC) stage. However, the aim of this study is the adjustments in the current SDLC and presented sustainable SDLC. In order to access the goal, the study firstly is used Standard Software development process method which is Capability Maturity Model Integration (CMMI) to institutionalization the model. Next, the research is used survey sample population is selected equal with 102 respondents that is Non-probability or not randomly obtained from international software organizations in Malaysia to identify main stages of the Green Software Development Life Cycle and finally is investigated based on that.
Keywords:	Software Development Life Cycle (SDLC), Green and sustainable software, Capability Maturity Model Integration (CMMI), Model
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	PREDICTION MODEL DEVELOPMENT FOR ONSHORE CRUDE OIL PRODUCTION BASED ON OFFSHORE PRODUCTION USING ARTIFICIAL NEURAL NETWORK
Author:	MARYONO, SUHARTONO
Abstract:	An old fact of the oil and gas industry is the difference in daily net production figures between onshore and offshore. This study aims to provide a model to reduce the difference, by processing offshore data to predict daily onshore production. The method used is regression and artificial neural network. The data used are daily data of offshore production and BSW (base sediment and water) number in 2013 until 2016. The results show that ANN shows better prediction compared with regression. From the observations made, the best period for modeling is the windowing of end period of 60 days with the tanh activation function and the standardized preprocessing method on ANN. For further research it is worth considering to include temperature as a prediction variable.
Keywords:	Crude Oil Production Figures, Prediction Model, Windowing, Regression, Artificial Neural Network.
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	IMPLEMENTATION OF A VEHICLE DETECTION SYSTEM IN THE FPGA EMBEDDED PLATFORM
Author:	ATIBI MOHAMED, BENRABH MOHAMED, ATOUF ISSAM, BOUSSAA MOHAMED, BENNIS ABDELLATIF
Abstract:	This document presents the implementation of a vehicle detection system in the FPGA platform. This system is based on two algorithms, an image processing algorithm that combines an algorithm for detecting areas of interest through the shadow of the vehicle, and an image descriptor (suspected zones) haar like features type, And another classification algorithm named artificial neural network which aims to detect the presence of the vehicles in these zones. To evaluate the results obtained, which showed that the proposed system is a fast and robust vehicle detector, a hardware implementation was performed in the FPGA embedded platform based on the NIOS II microprocessor and its input-output devices. The results of this implementation have shown that the FPGA platform has been able to maintain the performance of this system in terms of computational efficiency and speed, which confers on this system being a real-time vehicle detection system.
Keywords:	Vehicle Detection, Haar Like Features, Artificial Neural Network, FPGA, NIOS II, Real Time
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	AVERAGE AND MAXIMUM WEIGHTS IN WEIGHTED ROTATION- AND SCALE-INVARIANT LBP FOR CLASSIFICATION OF MANGO LEAVES
Author:	EKO PRASETYO , R. DIMAS ADITYO , NANIK SUCIATI , CHASTINE FATICHAH
Abstract:	The texture features would be important part when we conduct image classification. Local Binary Pattern (LBP) is one of feature extraction method that has most improvements by many researchers. Weighted Rotation- and Scale-invariant LBP (WRSI-LBP) is one of improvement versions. It uses minimum magnitude of local differences as an adaptive weight (WRSI-LBP-min) to adjust the contribution of LBP code in histogram calculation. The motivation is minimum magnitude gives minimum distortion to change LBP code in histogram calculation. In the classification of mango leaves case, the texture characteristic of mango leaves is highly difficult to be differed directly. So, for high accuracy detection, system requires texture feature with strength discrimination character, robust to illumination change, not sensitive to scaling and rotation. To achieve the goal, we propose average and maximum of magnitude of local differences as an adaptive weight of WRSI-LBP (WRSI-LBP-avg and WRSI-LBP-max). This scheme can be used to generate texture features for classification of mango leaves and general classification cases. The motivation of average weight is to cover all local different magnitude, because each LBP code generated would has unique neighbors pattern. The motivation of maximum is it gives maximum distortion to change LBP code, but it gives highest local different magnitude. We use Support Vector Machine (SVM) and K-Nearest Neighbor (K-NN) as classification methods. We use 240 images for performance evaluation, contains three varieties: Gadung, Jiwo and Manalagi. The K-Fold Cross Validation and Leave-One-Out are used as validation method. From the experiments show that WRSI-LBP-avg and WRSI-LBP-max achieve the highest accuracy compare to WRSI-LBP-min, LBP, Center Symmetric LBP (CS-LBP) and Dominant Rotated Local Binary Pattern (DRLBP). SVM achieve accuracy 75.21% with 16 bins, while K-NN achieve accuracy 79.17% with 256 bins. For uniform pattern, we apply experiments to WRSI-LBP-min, WRSI-LBP-avg, and WRSI-LBP-max. The highest accuracy is also achieved by WRSI-LBP-avg and WRSI-LBP-max.
Keywords:	Texture, Local Binary Pattern, Mango Leaves Classification, Rotation Invariant, Scale Invariant
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	SOURCE CODE ANALYSIS EXTRACTIVE APPROACH TO GENERATE TEXTUAL SUMMARY
Author:	KAREEM ABBAS DAWOOD, KHAIRONI YATIM SHARIF, KOH TIENG WEI
Abstract:	Nowadays, obtain program features becomes a hot issue in source code comprehension. A large amount of efforts spent on source code understanding and comprehension to develop or maintain it. As a matter of fact, developers need a solution to rapidly detect which program functional need to revise. Hence, many studies in this field are concentrating on text mining techniques to take out the data by source code analysis and generate a code summary. However, in this paper, we attempt to overcome this problem by propose a new approach (Abstract Syntax Tree with predefined natural language text Template (AST-W-PDT)) to generates human readable summaries for Java methods role. This paper describes how we developed a tool that the java source code can be summarized from the methods role. In evaluating our approach, we found that the automatically generated summary from a java class 1) is helpful to the developers in order to understand the role of the methods and will be useful, and 2) the automatically generated summary is precise.
Keywords:	Source Code Summarization, Program Comprehension, Source Code Maintenance, Abstract Syntax Tree
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	DEVELOPING OF THE CYBER SECURITY SYSTEM BASED ON CLUSTERING AND FORMATION OF CONTROL DEVIATION SIGNS
Author:	LAKHNO V. A., KRAVCHUK P. U, MALYUKOV V. P., DOMRACHEV V. N., MYRUTENKO L.V., PIVEN O. S.
Abstract:	The cyber security (CS) adaptive system is developed. It is based on advanced algorithms of anomalies signs space partitioning and attacks on clusters. A new approach of solving the topical scientific and applied problem of increasing the efficiency of systems of intelligent recognition of cyber attacks and anomalies is proposed. Unlike the existing ones, the present approach allows to take into account the modern statistical and remote parameters of the clustering of the attributes of cyber attacks and provides the opportunity to change the valid tolerance deviations for all the attributes simultaneously, as well as quickly identify new types of complex combined attacks with limited computing resources and variability of conditions. Unlike existing algorithms, the advanced ones enable to take into account the subject area peculiarities, including legal characteristics of cyber crimes of space signs construction. Among PTC Mathcad Prime 4.0, MATLAB (Simulink), using established simulations, the performance of the proposed algorithms are tested in CS systems of various companies.
Keywords:	Clustering Features, Cyber Security, Simulation Experiment, Test Tolerances
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	EXPERIMENTAL RESULTS ON MULTI-KEY SEARCHABLE ENCRYPTION TECHNIQUE WITH DIFFERENT ELLIPTIC CURVES AND APP DESIGNING
Author:	PUTTA SRIVANI, SIRANDAS RAMACHANDRAM, RANGU SRIDEVI
Abstract:	Multi-Key searchable encryption scheme is a technique implemented to perform the keyword search on cipher text. This scheme can be practically applied for client server applications to achieve data confidentiality and allows the server to perform the operation like search on the cipher data. Experimental results of multi-key searchable encryption scheme implemented for different types of elliptic curves are shown in this paper and results are compared between different types of elliptic curves. An application also designed with Java as frontend and MongoDB as backend. It also shows the search time taken to perform the search operation on the encrypted data by using this scheme.
Keywords:	Encryption, Search Token, Delta, Token, Cloud, Multi-key
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	EVALUATION OF THE IT OUTSOURCING PERFORMANCE IN THE DEVELOPMENT OF BIGDATA SYSTEMS
Author:	ELENA N. FOKINA, FEDOR YA. LEGOTIN, VERONIKA YU. CHERNOVA
Abstract:	The article considers the issue of outsourcing development in the field of information technologies. The work is devoted to the construction of a model that allows optimizing the organizational structure of the BigData information network and carrying out a comparative analysis of the effectiveness of its implementation in the format of IT outsourcing. In the process of research, the theoretical bases of outsourcing are analyzed and factors influencing the choice of the system implementation variant are generalized. In the critical review, a hypothetical assumption is made that the degree of IT outsourcing implementation effectiveness is inversely related to the degree of change in the characteristics of uncertainty, the frequency of queries and specificity. The study made it possible to characterize a set of indicators for evaluating the efficiency of outsourcing the BigData information system. The development model can be used to optimize the server complex of the network in order to reduce the number of cluster computing capacities involved during variable load. The criteria of the expediency of outsourcing application in practical activity of the companies are determined.
Keywords:	Bigdata, IT Outsourcing, Information Uncertainty, Information Network Development
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	MP4 VIDEO STEGANOGRAPHY USING LEAST SIGNIFICANT BIT (LSB) SUBSTITUTION AND ADVANCED ENCRYPTION STANDARD (AES)
Author:	PUTU ARI SRI LESTARI EKA NINGSIH, GUSTI MADE ARYA SASMITA, NI MADE IKA MARINI MANDENNI
Abstract:	Data transmission over the internet is vulnerable for attacked, therefore additional data security is required. Steganography is one of security protocol which used for protect the data. This research applied steganography using MP4 video as the host (cover media) and JPEG image as the embedded data. An AES-128 bit encryption was applied to the image before it embedded into the video. Encrypted image then stored bit-per-bit into an array. Those bits array then were embedded into the video by modifying LSB of the bytes of video. The image was embedded in the first still-encoded sample of the video. Sample was found by using atoms of MP4 container, those were stco/co64, stsc, stsz, stts dan mdat. Those atoms stored metadata of the video, thus could be used for mapping position of all the samples. The LSB substitution process did not touched the color spaces of the video, thus video quality can not measured using Peak Signal-to-Noise Ratio (PSNR) value. Video quality measured by a software called Qualify and the stego-video is scored 80 for its quality. That value indicated stego-video had an excellent quality. Stego-video could played well at all the common video player on any devices and there was no quality degradations or any changed as seen by human eyes. This research could contribute on increased the security of an image data when stored at the cloud storage or being transmitted over a network.
Keywords:	Steganography, Video, Sample, LSB, AES
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	VERIFICATION OF VHDL DESCRIPTIONS OF PARALLEL ARRAYS OF FINITE STATE MACHINES
Author:	NIKOLAY ALEKSANDROVICH AVDEEV , PETR NIKOLAEVICH BIBILO , VLADIMIR VLADIMIROVICH KOROBKIN , ANNA EVGENIEVNA KOLODENKOVA
Abstract:	Finite state machines are widely applied in the development of digital systems for description of control logic nodes, microprocessors, interface circuits and so on. This work proposes verification procedure of VHDL description of parallel arrays of finite state machines in Questa Sim simulation system. The main advantage of Questa Sim is that the model of finite state machine (FSM) can be verified if its written according to certain template. Verification is comprised of validating for compliance of VHDL description of finites state machine array with design specifications. The method utilizes the capabilities of the Questa Sim system, which makes it possible to identify the oriented graphs of the transitions of the component machines and to calculate the number of the arc passings in the graphs based on the results of simulation. However, the Questa Sim system does not recognize the FSM network and does not have the means to construct the tests based on the simulation results. Therefore, to solve these problems, it is suggested to store the simulation results, the sequence of input sets (stimuli) and the state tuples of the component machines, and to check the execution of transitions in the state graph of the machine network based on the sequences obtained, and, thus, to conduct the verification. In addition, this article discusses an example of description of FSM and FSM arrays using VHDL.
Keywords:	Digital Systems, VHDL Descriptions, Verification, Simulation, Finite State Machines.
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	AUTOMATIC CHRONOLOGICAL ORDERING OF AUDIO DATA USING SPECTOGRAMS
Author:	ALEXANDER ALFIMTSEV, SVETLANA NAZAROVA, Xiao Zelong
Abstract:	An automatic quantitative method for speech fragment analysis and chronological ordering is proposed. The method works by first converting the audio data into two-dimensional spectrograms and then extracting a large set of 1030 numerical descriptors (features) from the raw spectrograms, as well as from transforming the spectrograms. The audio fragments similarity value is computed using a variation of the Weighted K-Nearest Neighbour scheme. The similarity tree is then used to visualize the differences between the speech fragments. The accuracy of the method depends on the size of the feature set for the analysis and the length of the audio files. The speech fragments of the well-known politicians Vladimir Putin, Barak Obama, Angela Merkel, Jacques Chirac, George Bush and Vladimir Zhirinovsky were used for the analysis. The experimental results show that the method was able to create a chronological ordering of the speech fragments and that the most significant features for the analysis of audio data are: histograms of fuzzy-oriented gradients, multiscale histograms, and combinations of geometric moments.
Keywords:	Audio analysis, Chronological ordering, Fuzzy feature, Speech fragment, Two-dimensional spectrogram
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	MODELING PERFORMANCE MANAGEMENT OVER CORPORATE INFORMATION SYSTEM OPERABILITY
Author:	WALDEMAR WOJCIK, NAIZABAYEVA LYAZAT, ORAZBEKOV ZHASSULAN
Abstract:	The article herein solves the system task of controlling the distributed computer system operability. We are studying the speed performance and enterprises and organizations Corporate information system properties, representing complicated organizational technical complexes. Hereby there is considered the reaching the ways of intermittent and continuous speed monitoring and that of the stream volume, handled with Corporate information systems subsystems, modeling the current Corporate information systems components functionality. There is constructed the control conceptual model with the corporate information systems components functionality capacity, supporting the corporations business processes. There is presented the organizations corporate information system model. Organizations corporate information system possesses the possibility of the information systems content automatic adjustment to different users categories, being the representatives of the organizations various structural subdivisions.
Keywords:	Data Models, Corporate Information System, Business Process, Server, Database Control System (DCS)
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	THE DEVELOPMENT OF INTELLIGENT SYSTEMS FOR SOLAR PANELS STATION AND METHODS FOR DETERMINING THE PRECISION OF THE SOFTWARE TRACKING SYSTEMS , THE USE OF WIRELESS COMMUNICATION DEVICE
Author:	SATYBALDIYEVA F.A., BEYSEMBEKOVA R.N., SARYBAEV A.S., ESENBEKOVA G.J.
Abstract:	Use of wireless communication between detectors, transducers and industrial logical controllers in modern optical SPS heliostat control systems is more advantageous than laying hundreds meters of cable. To provide power supply, it is proposed to equip each heliostat with a self-contained power supply, since heliostat operates when concentrated solar radiation in the receiver is sufficient for steam generation, while the rest of the time it is in the standby mode. That is why use of a solar battery-powered self-contained power supply is more advantageous than use of centralized power supply from the industrial network. As distinct from other measuring informational systems, the described heliostat control system operates only when tracking parameters deviate towards the maximum permitted values. Presented in article the model of system of automatic control of the heliostats constitutes one of the varieties of measurement and information and control systems. Unlike other measurement systems described control system of the heliostats works only when the deviation of the tracking parameters in the direction of the maximum. The use of modern control systems, heliostats optical system SES wireless connections between sensors, transducers and industrial logic controllers is advantageous compared to laying hundreds of meters of cables.
Keywords:	Intelligent Control Systems, Mathematical Model,, Heliostat, Wireless Communication, Power Supply.
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	INNOVATIVE PARADIGM OF EDUCATION OF KNOWLEDGE COMPETENCY FORM BASED ON ONTOLOGY
Author:	KUBEKOV B.S., BEYER D., UTEGENOVA А.U., ZHAKSYBAEVA N.N.
Abstract:	The article describes an innovative methodology for shaping educational components of a planned training, based on ontology of reference concepts of educational content and an algebraic model of knowledge mapping - expression of knowledge on the basis of which subjects of the foundation and major-specific courses curriculum should be designed. The main purpose of the proposed methodology is to create knowledgeable content of the semantic database to design teaching and methodological materials for courses curriculum.
Keywords:	The Ontology Model, Characteristic Model, Concept, Specifying Concept, Consistency And Variability, A Composition Relations, Aggregation Relations, An Alternative Choice Relations, The Concept Of Paradigm, Specification Of Knowledge Expressions, Graph Models
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	STUDYING NETWORK TRAFFIC USING NONLINEAR DYNAMICS METHODS
Author:	SHARAFAT A. MIRZAKULOVA, VYACHESLAV P. SHUVALOV, ALEKSEY A. MEKLER
Abstract:	The development of the Internet, the constantly growing number of network users, and their mutual exchange of information is becoming an important communication bridge. However, this causes a series of technical difficulties, one of which is the growing requirements to network and server equipment and its maintenance. Therefore, the purpose of this study is to develop a computer program for teaching a neural network based on a computer network traffic table. A set of methods was used to achieve the set goal, including analysis, deterministic chaos, and systematization. The study used such software packages as TISEAN, MatLab, NetEmul, and Excel. The study generalized the experience that was relevant to the problem at hand. The study calculated the Lyapunov exponent, which characterizes the presence of chaos in the system. Analysis of the Lyapunov exponent enables using nonlinear dynamics methods to study the nature of the incoming and outbound traffic. With the help of the developed program, the neural network router is capable of prediction short-term parameters of a computer network; this information is sent to the system administrator, which will allow adapting the router to the estimated changes in the computer network.
Keywords:	Distribution series, Self-similarity, Chaotic processes, Phase portrait, Mutual information
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	SPAM DETECTION ISSUES AND SPAM IDENTIFICATION OF FAKE PROFILES ON SOCIAL NETWORKS
Author:	BALOGUN ABIODUN KAMORU, AZMI BIN JAAFAR OMAR, MARZANAH A.JABAR, MASRAH AZRIFAH AZMI MURAD, ABDULMAJID B. UMAR
Abstract:	Spam has been a major and global threat, Social networks have become our daily live and everyday tools, while different social networks have different target groups. With the rapid growth of social networks, people tend to misuse them for unethical and illegal conducts, fraud and phishing. Creation of a fake profile becomes such adversary effect which is difficult to identify without appropriate research. The current solutions that have been practically developed and theorized to solve this issue of spam detection issue and spam identification of fake profiles, primarily considered the characteristics and the social network ties of the user social profile. However, when it comes to social networks like Facebook, Twitter, SinaWeibo, Myspace, Tagged and LinkedIn such a behavioural observations are highly restrictive in publicly available profile data for the users by the privacy policies. The limited publicly available profile data of social networks makes it ineligible in applying the existing approaches and techniques in fake profile spam identification. Therefore, there is a need to conduct targeted research on identifying approaches for fake profile spam identification on selected and available data set of Facebook, Twitter and Sina weibo. In this research, we identify the minimal set of profile data that are necessary for identifying Fake profiles in Facebook, Twitter and Sina weibo and identifying the appropriate data mining approach and techniques for such task. We demonstrate that with limited profile data our approach can identify the fake profile with 84 % accuracy and 2.44 % false negative, which is comparable to the results obtained by other existing approaches based on the larger data set and more profile information.
Keywords:	Social Networks, Fake Profile, Spam Detection, Principle Component Analysis, Spam Identification
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	A RE-CONSTRUCTIVE ALGORITHM TO IMPROVE IMAGE RECOVERY IN COMPRESSED SENSING
Author:	R.JAYA LAKSHMI
Abstract:	In this paper, we study the Compressed Sensing (CS) image recovery problem. The traditional method divides the image into blocks and treats each block as an independent sub-CS recovery task. This often results in losing global structure of an image. In order to improve the CS recovery result, we propose a nonlocal estimation step after the initial CS recovery for de-noising purpose. The nonlocal estimation is based on the well-known nonlocal means (NL) filtering that takes advantage of self-similarity in images. We formulate the nonlocal estimation as the low-rank matrix approximation problem where the low-rank matrix is formed by the nonlocal similarity patches. An efficient algorithm, Extended NonLocal Douglas-Rachford (E-NDLR), based on Douglas-Rachford splitting is developed to solve this low-rank optimization problem constrained by the CS measurements. Experimental results demonstrate that the proposed E-NDLR algorithm achieves significant performance improvements over the state-of-the-art in CS image recovery.
Keywords:	Reconstruction, Algorithm, Recovery, Image, Compressed Sensing
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	LARGE SCALE SENSOR DATA PROCESSING BASED ON DEEP STACKED AUTOENCODER NETWORK
Author:	NAGWA ELARABY, MOHAMMED ELMOGY, SHERIF BARAKAT
Abstract:	Recently, Internet of Things (IoT) extremely populated by massive amounts of connected embedded devices, which are gathering large volumes of real-time heterogeneous data. Hence, IoT becomes an archetypal instance of Big Data. The collected IoT Big Data may not be profitable unless we evaluate and accurately exploit them. Providing mining for large scales of raw sensor data is an open challenge. To cope with this challenge, we proposed a system that operates in two modes, which are preparation and processing. The preparation mode converges on reducing the factors that hinder making efficient processing by focusing on three stages. First, handling missing data by applying interval-valued fuzzy-rough feature selection methodology. It highlights the most important features that contain missing data and gets rid of the others. Then, Maximum Likelihood (ML) approach is used for estimating the missing values. Second, anomalies are detected by initially utilizing K-nearest neighbors (KNN) algorithm then removing the detected ones from the data. Third, the dimensionality of nonlinearly separable data is reduced by exploiting Self-Organizing Map (SOM) network. In the processing mode, we passed the prepared data to a straightforward classifier based on a Deep Learning (DL) approach. We used autoencoder networks in constructing a deep network, which is the Deep Stacked Autoencoder (DSAE). The extracted features by the DSAE are non-handcrafted and task dependent, which gives it the most discriminative power to work as an efficient classifier. We apply the proposed model to PAMAP2 Physical Activity Monitoring data set. The results show that DSAE achieves high accuracy (99.8%) compared to the state-of-the-art classifiers.
Keywords:	Internet of Things (IoT), Big Data, Deep Learning (DL), Deep Stacked Autoencoder (DSAE)
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	CATEGORY EFFECT OF KEYWORDS ON HEALTHCARE INDUSTRY BY TEXT MINING IN WEB
Author:	JAE-WON HONG, JAE-YOUNG MOON
Abstract:	This study is intended to investigate the effects of keywords by category on the diffusion of consumers interest in healthcare industry focus on tourism using data mining in web environment. And the results are as follows: first, in the case of health care-related keywords, search activities for the categories of finance, biz, book, and social categories are found to affect the diffusion of medical tourism. Second, in the case of the physical examination keywords, search activities for the categories, such as Shopping, Travel, Social, community, are found to have some effect on the diffusion of medical tourism. Third, the role played by key words that affect the diffusion of medical tourism is appeared to differ depending on their categories. That is, for the part of health care, the categories of premeditation attribute, such as, finance, book, biz, travel, found to have impact on the diffusion of medical tourism, while in the case of physical examination, categories of impulsion, such as shopping, sports, food, are found to affect the diffusion of medical tourism. In other words, it is implied that the attributes of keywords and categories should be considered in communicating for the diffusion of medical tourism in web environment.
Keywords:	Keyword Analysis, Text Mining, Healthcare, Tourism, Web Environment
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	THE STUDY ON THE EFFECT OF CONTROL FACTORS FOR JOINT VENTURE SHARING INFORMATION USING ENTERPRISE RESOURSE PLANNING SYSTEM
Author:	HYO-KYUNG KIM, WON-HEE LEE
Abstract:	This empirical study is to investigate the effects of commitments (calculative commitment, affective commitment) and control types (output, process, and social), which are relational characteristics between partners in international joint venture, on corporate performance. Today almost companies have adaptive their own ERP system. When companies which using ERP system want to delivery information from cooperate companies without information bias. Especially, companies which have cooperate companies concern about information quality because company send misinformation to cooperate companies, they cannot make the right decision. Joint venture is popularly spread in the world. Sharing information is really important thus many joint ventures have built their local ERP system and sharing information each others. According to the results of this research, affective commitment has positive effect on all the output control, process control, and social control. Whereas, calculative commitment has positive effect on the output control and process control only. But neither output control nor process control has positive effect on the results
Keywords:	ERP System, Information Bias, Joint Ventures, Information Quality Assurance, Information Sharing, Affective Commitment, Calculative Commitment
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	IMPACT OF DEFAULT OPTION ON FINAL PRICE IN ONLINE COMMERCE
Author:	JUN SIK KWAK
Abstract:	As online purchases become more frequent, the e-commerce and m-commerce market is growing rapidly. Under such circumstances, understanding the preferences of online consumers is very important. Status quo bias and implicit self-theory of people can give important implications for online purchase. People do not want to change the current state. So, people prefer to maintain their current state rather than change. This person's tendency is called the status quo bias. A typical strategy that utilizes the tendency of people who do not like change is the default option strategy. Default option means an option that is selected automatically unless an alternative is specified. Online commerce experiment was conducted to study default option effect in final purchase price. There are two types of default option. One is the additive default option that people add their desired items with basic item. The other is subtractive default option that people delete items they dont want with full items. The results show that the final purchase price is higher in subtractive default option than in additive default option. The final purchase price is different from sex. The final purchase price of male group is not different from between additive default option and subtractive default option. But the final purchase price of female group is higher in subtractive default option than in additive default option. This result shows that female feels more losses than male. Implicit self-theory influence on the final purchase price in default option. Especially, in the subtractive default option, the final price of incremental theorist is higher than that of entity theorist. Therefore, E-commerce and M-commerce companies will have to establish effective online sales strategies that take into account default options and implicit self-theory.
Keywords:	Online Purchase Psychology, Status Quo Bias, Default Option, Implicit Self-Theory, Final Price
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text

Title:	TEXTURE RECOGNITION USING CO-OCCURRENCE MATRIX FEATURES AND NEURAL NETWORK
Author:	SUHAIR H. S. AL-KILIDAR, LOAY E. GEORGE
Abstract:	Texture Recognition is used in various pattern recognition applications and classification texture that possess a characteristic appearance. This research paper aims to provide an improved scheme to provide enhanced classification decision with need to increase the procession time significantly. This research studied the discriminating characteristics of textures by extracting them from various texture images using Gray Level Co-occurrence Matrix (GLCM). Three different sets of features are proposed: the first set is a simple modified features extracted from the traditional Gray Level Co-occurrence Matrix (GLCM); the second set uses two sub-sets of features extracted from two GLCM calculated using two displacement values; the third way was passing the extracted set of GLCM through Artificial Neural Network (ANN) for classification purpose. The considered method was applied on 13 classes of textures belong to three sets from Salzburg Texture Image Database (i.e., bark, marble and woven fabric), each set holds 16 images per class, so the total 208 image images were tested. Each image was separated into four bands of color component (i.e., red, green, blue, and gray). So, the analysis work was done for 56 different set of characteristics (14 features per band). Then calculating the average and standard deviation and finally conducting scatter analysis for each feature to find out the best discriminating features which can be used for attaining best classification. The experiments showed that the results are competitive, the test results indicated that the attained average accuracy of classification is improved up to (99.71%) for the training set and (99.3 %) for the testing set.
Keywords:	Texture Recognition, characteristics, Co-occurrence Matrix (GLCM) features, Artificial Neural Network (ANN), Features extraction.
Source:	Journal of Theoretical and Applied Information Technology 15^th November 2017 -- Vol. 95. No. 21 -- 2017
Full Text