		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 2015 | Vol. 81 No.1

Title:	IMPLEMENTATION IN DECISION SUPPORT SYSTEM OF ONTOLOGICAL APPROACH TO EXTRACTION OF DATA ABOUT INFORMATION OBJECTS FROM NEWS FLOW
Author:	ALEKSANDR VALEREVICH VOLKOV, DMITRII ALEKSANDROVICH SYTNIK
Abstract:	The objective of this article is the study of existing approaches to extraction and structuring of data about information objects from news flow and development of the original approach to solution of that task. The article covers the approach to extraction of data about information objects based on domain otology. The developed model of information objects extracted from text is described, as well as the complex of linguistic resources applied for implementation of extraction from text of data about various types of information objects. This article gives a brief description of processing. The stage of extraction of data about information objects includes three phases: named entity recognition; inter-object relationship extraction; building complex information objects (events). For extraction of information from text, rule-based approach is used.
Keywords:	Ontology, Text Extraction, Information Extraction (IE), Rule-Based Extraction, Event Extraction, Natural Language Processing, Unstructured Data, Unformalized Data, Decision Support System, News Processing, Analysis Of News Flows, Mass Media Monitoring.
Source:	Journal of Theoretical and Applied Information Technology 10^th November 2015 -- Vol. 81. No. 1 -- 2015
Full Text

Title:	ALGORITHMIZATION OF SEARCH OPERATIONS IN MULTIAGENT INFORMATION-ANALYTICAL SYSTEMS
Author:	ANASTASIA GENNADIEVNA ANANIEVA, ALEXEY ANATOLIEVICH ARTAMONOV, ILYA URIEVICH GALIN, EVHENIY SERGEYEVICH TRETYAKOV, DMITRY OLEGOVICH KSHNYAKOV
Abstract:	The exponential growth in the number of data sources and data on the Internet leads to necessary selection of optimal searching engines for data collection that corresponds to user’s request. The task to create searching algorithms lies primarily in the decomposition of data types, selection of appropriate information sources and adjustment of searching engines (usage of specific syntax) for collection of large amounts of data from reliable sources. Since 2008 at the department “Analysis of competitive systems” NRNU MEPhI are held activities on setting up specific agent-based searching systems and creation of unique searching algorithms and techniques for analyzing the output technological developments related to advances in the area of scientific and technological sciences. In particular, the resource cluster was collected and thesauruses were compiled on such advanced scientific fields as “Plasma physics”, “Nanotechnology” and “Laser technology”.This article describes the algorithms, approaches and methods of data search retrieval from the Internet that are developing and testing at the department №65 “Analysis of competitive systems” NRNU MEPhI under the “multi-agent information and analytical systems in the area of scientific and technology science” MIAS [1].
Keywords:	Multi-agent system, Big Data, Data mining, retrieval system, information and analytical system
Source:	Journal of Theoretical and Applied Information Technology 10^th November 2015 -- Vol. 81. No. 1 -- 2015
Full Text

Title:	DEVELOPMENT OF THE MATHEMATICAL MODEL FOR ANALYSIS OF LEVELING POINTS STABILITY OF REFERENCE LEVELING GRID
Author:	IBRAGIM GILANIEVICH GAIRABEKOV, HASAN ELIMSULTANOVICH TAYMASKHANOV, MAGOMED SALAMUVICH SAYDUMOV, SHARPUTDI SHAMSUTDINOVICH ZAURBEKOV, BERS MOVSAROVICH KHASAEV, MAGOMED SHAVALOVICH MINTСAEV
Abstract:	The main drawback of the current methods used for analyzing stability of leveling points in the reference leveling grid have been defined. The problem of developing a mathematical model of leveling points in the reference leveling grid is being solved. Equal accuracy of measurement in different cycles is expected. Possible subsidences between the reference points are represented by a polynomial of the second degree. In solving this problem, the polynomial coefficients are calculated using the least squares method. Accuracy of the obtained mathematical model used to make a conclusion about the need to identify elevations between the reference points with high accuracy (up to tenths of a millimeter) has been estimated.
Keywords:	Hyperbolic Equation, Nonlocal Integral Condition, Bessel Operator, Mathematical Model, Reference Point, Leveling Grid, Elevation, Subsidence, Polynomial, Differentiation, Model Error
Source:	Journal of Theoretical and Applied Information Technology 10^th November 2015 -- Vol. 81. No. 1 -- 2015
Full Text

Title:	APPROACH TO THE ASSESSMENT OF UNIVERSITY GLOBALIZATION INDEX
Author:	MAXIM V. SERGIEVSKIY, KONSTANTIN S. ZAYTSEV
Abstract:	It is impossible to imagine the present-day world without changes of politics, economics, social life caused by globalization. The globalization within society influences and updates the system of education quite naturally that is a factor of competitiveness advance and of the modern nation development. That is why in order to determine the general level of nation globalization it is necessary to determine the globalization level of education which is a boundary zone between social life and culture. The purpose of this article is to analyze the informative aspect of factors influencing on integration processes of university education and to formulate an approach for quantitative estimation of globalization index of education of a separate university and of a set of universities. Quality of education as well as its involvement in world global processes is mainly determined by curriculum content and first of all by the set of subjects composing this curriculum. The absence of subjects that are of great importance for the particular scientific sphere is this very factor that reduces the quality of education. To a lesser degree the quality of education also depends on the way the academic activity is organized including teachers’ qualifications and the way classes are held. That is the reason why at this work we describe an approach to the evaluation of the globalization level of a separate university, which is based on the comparative analysis of its curriculum content of different subjects with the curriculum content of the world leading universities, or the so-called “ideal curriculum”. In order to calculate the globalization index of education in the informative aspect we have introduced a relative additive criterion that uses weighted values of subjects included in the overall ranking.
Keywords:	The Globalization Index Education, Globalization Systems Education, Modernization Of Education, Quality Of Education, Content Of Education, Curriculum, Information Technology
Source:	Journal of Theoretical and Applied Information Technology 10^th November 2015 -- Vol. 81. No. 1 -- 2015
Full Text

Title:	THE EXPERIMENTAL STUDIES OF THE JPWL TOOLS ABILITY TO CORRECT BURST ERRORS IN A NOISY CHANNEL WHEN TRANSMITTING VIDEO IN A JPEG 2000 FORMAT
Author:	SERGEY VASILIEVICH SKOROKHOD
Abstract:	The discussion is made on the issue of transmitting video in a JPEG 2000 format using the JPWL tools (Wireless JPEG 2000) for noise-immunity coding through a noisy channel under conditions of burst errors occurrence. The major problem is the restoration of the lost packet using the JPWL tools. It is set out to conduct an experimental study on the ability of JPWL tools to correct burst errors. The methodology of the study consists in modelling a transmission of the JPWL protected video composed of 1,000 frames of one and the same image with a size of 1,024 x 768 pixels. The variable parameters of the study are as follows: the Reed-Solomon codes applied for protection, the number of image tiles, and the RTP (Real-time Transport Protocol) packet loss ratio. A standard variant of protection and a combination of the standard variant with an interleaving algorithm are considered. The final results of the experiment are the average values of the percentage of fully and partially restored tiles of the number of tiles in a code stream. A software system developed for the conduction of study is described, which includes the JPWL encoder and decoder, the tools for partitioning code streams into RTP packets, the tools for packet loss modelling, and the tools for frame assembly from the RTP packets. A macro flowchart of the JPWL encoder and decoder functioning is described. A method for the code stream interleaving is suggested, which is intended for increasing its resistance to burst errors. During the study process, the two hypotheses were put forward, which were confirmed by the experimental results obtained. The first hypothesis is that the standard JPWL tools are not able to correct burst errors. The second hypothesis is that the standard JPWL tools used in conjunction with the interleaving algorithm can recover burst errors. A side effect of the study is the conclusion on a slight improvement of the recoverability of tiles with the increasing number of tiles in a code stream.
Keywords:	JPEG 2000, JPWL, Burst Errors, Noise-Immunity Coding
Source:	Journal of Theoretical and Applied Information Technology 10^th November 2015 -- Vol. 81. No. 1 -- 2015
Full Text

Title:	CROSS-RELATIONAL STUDY BETWEEN INTELLIGENCE AND BRAIN ASYMMETRY ABILITIES USING EEG-BASED IQ CLASSIFICATION MODEL
Author:	AISYAH HARTINI JAHIDIN, MEGAT SYAHIRUL AMIN MEGAT ALI, MOHD NASIR TAIB, NOORITAWATI MD. TAHIR, AHMAD IHSAN MOHD YASSIN
Abstract:	EEG is an established technique that has been widely implemented in brain research. Recent innovations in signal processing approaches have allowed implementation of EEG beyond the clinical settings. It has been widely acknowledged that each individual exhibits multi-facet potential which is rooted in the brain. Hence, there is a possibility that these abilities are inter-related with each other. This study proposes to map the relationship between intelligence quotient (IQ) and brain asymmetry (BA) using EEG and intelligent signal processing approach. EEG-based ANN modelling has been previously developed and enhanced. The model is then implemented to predict three levels of IQ (low, medium and high IQ) from 51 samples with distinct BA indices. The indices are derived from Alpha asymmetry score. Mapping between IQ and BA revealed that through a systematic technique; significant relationship exists between distinct IQ levels and symmetrical brain indices. Findings have demonstrated that the best performance occur when subjects maintain relatively balanced control between the two hemispheres. Hence, by implementing feedforward neural network model based on EEG power ratio features, the attempt to relate IQ with BA has been realised with promising results.
Keywords:	Artificial Neural Network (ANN), Brain Asymmetry (BA), Electroencephalogram (EEG), Intelligence Quotient (IQ), Power Ratio
Source:	Journal of Theoretical and Applied Information Technology 10^th November 2015 -- Vol. 81. No. 1 -- 2015
Full Text

Title:	AUTHENTICATION MECHANISM FOR CLOUD NETWORK AND ITS FITNESS WITH QUANTUM KEY DISTRIBUTION PROTOCOL: A SURVEY
Author:	ROSZELINDA KHALID, ZURIATI AHMAD ZUKARNAIN, ZURINA MOHD HANAPI, MOHAMAD AFENDEE MOHAMED
Abstract:	Communication in a huge network such as in cloud infrastructure is vast in demand. Almost all classified information transferred via the communication channel. Most types of attack can cause the classified information be in the hand of unauthorized party. This situation will lead to information disclosure and the user feel unsecured for using the services like offered by cloud infrastructure. Besides, we look into authentication as a primary concern; we also pay attention to the secure communication channel. We believe, with a secure communication channel additional with a secure authentication we may reduce the possibility of attack that may lead to information disclosure. This paper summarizing the issue and challenge facing in cloud authentication mechanism. We also review existing technique of authentication mechanism for cloud network and discuss important issues in this field such threat and insecure scheme, whereby the root of these kinds of issue originates from the weak-ness of the authentication mechanism on the security channel being using. After all, we found the implementation of quantum key distribution scheme in an enormous network such as cloud infrastructure may resolve the issue interception of unauthorized party in the network. A study investigating how the use of quantum key distribution key, can guarantee that the message has not been modified or replaced by a dishonest party with control of the communication line.
Keywords:	Cloud Authentication, Multi-Party Quantum Key Distribution, Secure Communication
Source:	Journal of Theoretical and Applied Information Technology 10^th November 2015 -- Vol. 81. No. 1 -- 2015
Full Text

Title:	NEW MODEL OF FRAMEWORK FOR TASK SCHEDULING BASED ON MOBILE AGENTS
Author:	YOUNES HAJOUI, MOHAMED YOUSSFI, OMAR BOUATTANE, ELHOCEIN ILLOUSSAMEN
Abstract:	Compute-intensive applications and applications with high volume of data need strong processing power and considerable storage resources. To reach the required performance, multiple machines should be associated in order to handle the distributed tasks. In this paper, we propose a new framework for task distribution based on mobile agents. In the proposed model, a dispatcher agent is used to distribute parallel tasks to worker agents. Each worker agent is deployed in a node of the distributed system according to the load balancing system. The proposed framework is build using three layers which are the user task producer, the scheduling load balancing layer and the workers layer. After presenting the architecture and the structure of the proposed model, an example of application, relating to the distributed image processing, is presented to improve the performance of this framework.
Keywords:	Task scheduling; Parallel Computing; Distributed System; Framework; Multi-Agent system; Load balancing.
Source:	Journal of Theoretical and Applied Information Technology 10^th November 2015 -- Vol. 81. No. 1 -- 2015
Full Text

Title:	AUTOMATED TRANSFORMATION APPROACH FROM USER REQUIREMENT TO BEHAVIOR DESIGN
Author:	NURI JAZULI KAMARUDIN, NOR FAZLIDA MOHD SANI, RODZIAH ATAN
Abstract:	System design is an important process in the development of any type of computer related system and a process of gathering user requirement of the system that will be developed. System design is important in software development because it assists the developer to develop a system according to what are required by the user. One of main components in system design is Unified Modeling Language (UML) diagram. UML diagram is use as a model to show all important functions, process, flows, actors, classes that related to the system that will be built. This research is focusing on two types of diagram from behavior diagram which are Use case diagram and Activity diagram. Many developers have come out with a tool to help the analyst to draw the diagrams in the computer. Tools like Rational Rose, Lucidchart, UMLet are an example of the tool that have been developed. This step is still not preferred by system analyst because drawing use case diagram and activity diagram takes time and it must be done manually. Model Transformations is one of new finding in system design that intended to ease system analyst in modeling system design. This process is very useful in helping the analyst to reduce time in drawing use case and activity diagram. In this paper, we propose an approach to automatically transform user requirement into behavior model (Use Case diagram, Activity Diagram) and to develop a tool that will enable the transformation of the requirement into behavior model.
Keywords:	Model transformation (MT), UML model, System Design
Source:	Journal of Theoretical and Applied Information Technology 10^th November 2015 -- Vol. 81. No. 1 -- 2015
Full Text

Title:	LOCAL BINARY PATTERNS AND MODIFIED RED CHANNEL FOR OPTIC DISC SEGMENTATION
Author:	NUR AYUNI MOHAMED, MOHD ASYRAF ZULKIFLEY, AINI HUSSAIN, AOUACHE MUSTAPHA
Abstract:	Glaucoma is one of the ocular eye diseases that can cause gradual vision loss and permanent blindness if it is not treated in the early stage. Current screening test such as intraocular pressure (IOP) assessment is not efficient since eye pressure is not the only symptom of glaucoma. The most suitable assessment of the glaucoma is by analyzing the health of the optic nerve head. In order to quantify the severity level of glaucoma, an automated detection system is developed by examining the optic disc and optic cup size. This paper explores two methods for optic disc segmentation, a part of modules in automated detection system, which are local binary patterns (LBP) and modified red channel (MRC). Both methods utilized only the red channel of RGB format fundus image as it alone is enough to achieve good performance in term of image contrast as compared to the other channels. For each method, preprocessing is first performed to enhance the quality of the input fundus image and post-processing is performed to smooth out the segmented boundary of the optic disc. RIM-One database is used to validate the simulation results for both tested methods. The results show that MRC performance is more stable in wider conditions compared to LBP. In conclusion, both methods segment the optic disc boundary with high accuracy, which can be used to calculate cup-to-disc ratio to determine severity of glaucoma.
Keywords:	Local Binary Pattern, Textural Classification, Glaucoma, Fundus Image and Disc Segmentation
Source:	Journal of Theoretical and Applied Information Technology 10^th November 2015 -- Vol. 81. No. 1 -- 2015
Full Text

Title:	A REVIEW ON TOOLS OF RISK MITIGATION FOR INFORMATION TECHNOLOGY MANAGEMENT
Author:	BOKOLO ANTHONY JNR, NORAINI CHE PA
Abstract:	Information technology (IT) industries are facing operational. Technical and strategic risks that make IT practitioners miss their planned schedule, time and quality. Hence, there is the need to effectively and efficiently mitigate such risks if staffs want to avoid the above problems. Many studies have been accompanied for viewing the issues from different aspects. However, available IT risk mitigation tools present many weaknesses and above all, they are few. 13 studies related to the risk management and mitigation in IT has been reviewed. The studies have been reviewed based on the review method which is called Systematic Literature Review (SLR). Based on this approach, all previous studies related to this title have been studied systematically. Thus this paper aims to do review existing risk tool from 2000 until 2015. Also, this paper presents the results of the systematic reviews on the number of tools in IT risk mitigation. The findings of this review indicate that existing tools for risk mitigation are not completely effective and efficient in mitigating risk that occurs in IT management. This review identifies existing risk mitigation prototypes tools purpose, functionalities, architecture and limitations. The findings of this review also discuss the purpose, functionalities, architecture, limitations of the existing risk mitigation tool in IT Management.
Keywords:	Risk Mitigation, Risk Management, IT Management, IT Risk Tools, Risk Assessment Tools, Systematic Literature Review
Source:	Journal of Theoretical and Applied Information Technology 10^th November 2015 -- Vol. 81. No. 1 -- 2015
Full Text

Title:	GEOMETRY ALGORITHM ON SKELETON IMAGE BASED SEMAPHORE GESTURE RECOGNITION
Author:	AERI RACHMAD, MUHAMMAD FUAD
Abstract:	Semaphore, a way of communicating remotely, usually practiced in scouting activities. Information is delivered by gestures or movements using specific tools such as flags, paddles or rods. Teacher and instructors are needed for learning semaphore in conventional way as they will give examples and make correction when such an error occured. Based on the practical need to provide an alternative way to learn semaphore, this research proposes the use of geometry algorithm to develop a semaphore gesture recognition based on skeleton images that read from Kinect sensor. Euclidean distance and law cosines are two formulas that applied to generate gesture parameters of each alphabet. Recognition is achieved by comparing a pair of values of model and real-time gesture. Accuracy of this system that have been measured using RMSE with 30° of tolerance yields 90.76% for Alphabet and 88% for Word.
Keywords:	Semaphore Gesture, Geometry Algorithm, Kinect, Skeleton Image, Alphabet
Source:	Journal of Theoretical and Applied Information Technology 10^th November 2015 -- Vol. 81. No. 1 -- 2015
Full Text

Title:	INTELLIGENT DECISION SUPPORT SYSTEMS (IDSS) FOR MULTI-OBJECTIVE OPTIMIZATION PROBLEMS AT SEA SECURITY INDONESIA
Author:	MASROERI
Abstract:	Indonesian maritime security issues is a complex problem therefore included in the multi-objective problems, it is necessary for the proper method to overcome these problems, one of them by developing intelligent decision support system using methods Non dominated Sorting Genetic Algorithms II (NSGA II) and Fuzzy C Means (FCM). NSGA-II method will produce a set of optimal solution candidates, while the FCM method serves to minimize the set of optimal solutions. His study aims to maximize the achievements of the coverage area and minimize operational costs by considering the type of vessel, speed, range radar, endurance. Optimization results using NSGA-II with the following parameters: population = 20, generation = 100, crossover probability = 75% and = 90% probability of mutations produced as many as 125 candidates for the optimal solution. Furthermore, the solution is in the cluster to minimize the prospective solutions based on a predetermined point cluster into four, resulting in four candidates for the best solution, that is: [1] Co = Rp. 4.741.205.798 and Ca = 1.915.083 Mil2; [2] Co = Rp. 3.997.582.228 and Ca = 1.560.672 Mil2; [3] Co = Rp. 4.802.314.832 and Ca = 1.962.895 Mil2; [4] Co = Rp. 4.939.637.487 and Ca = 1.982.564 Mil2. Results of Intelligent Decision Support System with consideration of increasing security in sea area of Indonesia and optimize the budget that has been set by the Government, the recommended solution is the solution to the cluster of four with a combined value of patrol boats [647755475625136306153116502] Cost = Rp. 4.939.637.487, Coverage = 1.982.564 Mil2 and distance = 0.006724.
Keywords:	IDSS, Multi-Objective Optimization Problems, NSGA-II, FCM
Source:	Journal of Theoretical and Applied Information Technology 10^th November 2015 -- Vol. 81. No. 1 -- 2015
Full Text

Title:	MULTI-BAND POWER SYSTEM STABILIZER MODEL FOR POWER FLOW OPTIMIZATION IN ORDER TO IMPROVE POWER SYSTEM STABILITY
Author:	AGUS JAMAL, SLAMET SURIPTO, RAMADONI SYAHPUTRA
Abstract:	This paper presents a Multi-band Power System Stabilizer Model for power flow optimization in order to improve power system stability. Power System Stabilizer (PSS) is equipment that can be used to enhance the damping of power system during low frequency oscillations. For large scale power systems comprising of many interconnected machines, the PSS parameter tuning is a complex exercise due to the presence of several poorly damped modes of oscillation. The problem is further being complicated by continuous varied in power system operating conditions. In the simultaneous tuning approach, exhaustive computational tools are required to obtain optimum parameter settings for the PSS, while in the case of sequential tuning, although the computational load is fewer, evaluating the tuning sequence is an additional requirement. There is a further problem of eigenvalue drift. This paper presents the multi-band PSS model for designing robust power system stabilizers for a multi machine system. Simulations were carried out using several fault tests at transmission line on a Two-Area Multi-machine Power System. As a reference the PSS model, Delta w PSS and Delta Pa PSS has been used for comparison with the PSS under considerations. The result shows that power transfer response using the model is more robust than Delta w PSS and Delta Pa PSS, especially for three phase faults and phase to ground faults.
Keywords:	Power System Stabilizer, Multi-Band, Transient Stability, Oscillation, Multi-Machine Power System.
Source:	Journal of Theoretical and Applied Information Technology 10^th November 2015 -- Vol. 81. No. 1 -- 2015
Full Text

Title:	ATTENUATION CORRECTION OF PET IMAGE RECONSTRUCTED BASED ON DIRECT FILTERING OF THE RAW DATA ACQUIRED USING MCNPX CODE
Author:	M. SAEED, T. EL KHOUKHI, Y. BOULAICH, E. CHAKIR, H. BOUKHAL, T. EL BARDOUNI
Abstract:	This paper presents the Monte Carlo simulation of the Positron Emission Tomography (PET) scanning technique using MCNPX code. The raw data generated by general MCNPX code contain a lot of unnecessary information for the image reconstruction process. This requires a large memory to store all this information. In order to reduce the needed memory, we introduce some modifications on the MCNPX code source in order to be able to write directly the important data representing the coincidence events detected by the detector pairs along the lines of response (LORs). These data are acquired as a result of launching two simulations on a uniform cylindrical positron emitter source in water-filled environment. Thereafter, from the object’s geometry, knowledge of the materials in this object and the acquired simulation data, the attenuation correction map has been calculated and applied to the acquired data during PET image reconstruction process.
Keywords:	PET, MCNPX, Image reconstruction, Attenuation correction.
Source:	Journal of Theoretical and Applied Information Technology 10^th November 2015 -- Vol. 81. No. 1 -- 2015
Full Text

Title:	WEB-BASED GIS MAPPING FOR GEOTHERMAL RESOURCES POTENTIAL IN WEST SUMATRA
Author:	YUHENDRA
Abstract:	Geothermal energy is one of the renewable power sources which can be an alternative to the increasingly more scarce fossil fuel. Especially, West-Sumatera is one of region in Indonesia located at the ring of fire, has an abundant potential for geothermal energies. The proposed of research address was development of web-based mapping application and identifying potential areas for geothermal exploration. We use web-based GIS (also called internet GIS or on line GIS) application for standards software packages to develop an interactive web mapping portal for spatial analysis. The resulting system as for resources information system tools in decision support system (DSS) to helping and managing the company’s geothermal resources for key decision making, planning and also provides the users with an innovative and interactive way to access the spatial content over the internet/intranet. The service application provides a variety of functionalities such as; browse map mode, querying, navigation system, user requiring service, search region and displaying geothermal well properties e.g. megawatt capacity, map editing and printing from the web interface.
Keywords:	West-Sumatra, Mapping, Web-based GIS, Decision Support System, Geothermal Energy, Resources Information System
Source:	Journal of Theoretical and Applied Information Technology 10^th November 2015 -- Vol. 81. No. 1 -- 2015
Full Text

Title:	AN EFFICIENT GENTLE ADABOOST-BASED APPROACH FOR MAMMOGRAMS CLASSIFICATION
Author:	NEZHA HAMDI , KHALID AUHMANI , MOHA M’RABET HASSANI , OMAR ELKHARKI
Abstract:	In this work we have proposed and evaluated a new approach for classification of mammograms. This approach is based on Gentle AdaBoost. Our main contribution in this proposition is that the strong classifier is constructed by wheited weak classifiers. These weak classifiers are extracted from the sub-bands of discrete wavelet transform. We have used the Receiver Operating Curves (ROC) tool to evaluate our proposition with mammograms of MIAS Database. We have presented True positive rate versus false positive rate for different features types and Gentle AdaBoost iteration number. Obtained results show that the best area under curve (AUC), which represents the approach performance, is reached for Zernike moments and it is equal to 0,98 and 0.99 for T=10 and T=50 respectively. The other features types allow an AUC comprised between 0.7 and 0.76. Results show also that performance of the proposed approach is slightly improved if the number of iterations is increased.
Keywords:	Machine learning, Classification, Gentle AdaBoost, Discrete wavelet transform, ROC, Feature Extraction.
Source:	Journal of Theoretical and Applied Information Technology 10^th November 2015 -- Vol. 81. No. 1 -- 2015
Full Text

Title:	COMPARISON OF SOFTWARE RELIABILITY ANALYSIS FOR BURR DISTRIBUTION
Author:	Dr. G. SRIDEVI, Dr. C.M. SHEELA RANI
Abstract:	This model is for Burr type distribution with three parameters which is discussed in two versions - the Burr type III and Burr Type XII. In this paper, we compare the performance of two versions of the suggested model is tested on five real time software failure data sets. The versions perform with variable accuracy, which suggest that no universal “best” among the two versions of the model could be attained.
Keywords:	Burr type III Model; Burr type XII Model; NHPP; ML Estimation
Source:	Journal of Theoretical and Applied Information Technology 10^th November 2015 -- Vol. 81. No. 1 -- 2015
Full Text

Title:	CONCRETE CRACK DETECTION BASED MULTI-BLOCK CLBP FEATURES AND SVM CLASSIFIER
Author:	RGUIG MUSTAFA, EL AROUSSI MOHAMED
Abstract:	Recently Automatic concrete crack detection has been converted to a real challenge for high performance of the inspection and diagnosis of concrete structures images. Generally, there are various noises such as irregularly illuminated conditions, shading and divots in the concrete images. Hence it is difficult to detect cracks automatically. In this paper, a novel and efficient approach based on Compound Local Binary Pattern (CLBP) using support vector machines is proposed for automatic concrete crack detection. The contributions of this paper include the following steps: (1) the proposed system starts by pre-processing the database images, smoothing their texture and enhancing any existing cracks, being followed by the extraction of descriptive features. Here each image is divided into several non-overlapping blocks and each block originates a feature vector. (2) The support vector machine (SVM) is successfully applied to determine the concrete crack image classification. The experimental results gave a 97.43% classification accuracy rate, which indicate that the proposed method is a promising tool for analysis of concrete structures images.
Keywords:	Concrete Crack Detection, Concrete Structure, Compound Local Binary Pattern, Support Vector Machine, Feature Extraction, Local Binary Pattern.
Source:	Journal of Theoretical and Applied Information Technology 10^th November 2015 -- Vol. 81. No. 1 -- 2015
Full Text

Title:	FUZZY KERNEL C-MEANS ALGORITHM FOR INTRUSION DETECTION SYSTEMS
Author:	ZUHERMAN RUSTAM, AINI SURI TALITA
Abstract:	Intrusion Detection Systems (IDS) are used as security management systems. There are two approaches of IDS, Misuse Detection (knowledge-based intrusion detection) and Anomaly Detection (behavior-based intrusion detection). Misuse detection is performed by monitoring activities which is suspected as an intrusion based on prior information about specific attacks. While anomaly detection is based on the observation of the activity that is incompatible with the acceptable behaviors in normal conditions and makes it possible to determine new type of attacks in the system. Some Computational Intelligence models have been developed to solve Intrusion Detection Systems problems such as Neural Network and Neuro-Fuzzy methods. They are chosen because IDS involves large data sets with several different features that can bring out negative effects on IDS accuracy and its computational time. Naïve Bayes, Decision Tree (C4.5) and Kernel Matrix Methods can be used to reduce the number of features at data sets. We propose Fuzzy Kernel C-Means Algorithm as another method to solve IDS problems that we claim provides better results while combined with Kernel Matrix method to reduce the number of selected data features.
Keywords:	Data Features, Fuzzy C-Means, Intrusion Detection Systems, Kernel Matrix, Kernel Method
Source:	Journal of Theoretical and Applied Information Technology 10^th November 2015 -- Vol. 81. No. 1 -- 2015
Full Text