|
Research Institute of Intelligent Computer Systems Ternopil National Economic University |
|
2008, Vol. 7, Issue 1 |
|
Contents and abstracts
CLASSIFICATION DATA EXPLORATION METHODS IN MODERN REALTIME DATA WAREHOUSE Jakub Chłapiński, Piotr Mazur, Jan Murlewski, Marek Kamiński, Bartosz Sakowicz
Department of Microelectronics and Computer Science, The goal of this article is to introduce problems that may arise during analysis of classification methods used in data mining applications. In the following sections some of the most common classification techniques are described along with several proposed extensions which allow these methods to be used in incremental data warehouses. The primary focus was aimed at the problem of performing incremental learning methods that may be used in near realtime data warehousing applications. VISUAL PATTERN STRUCTURE INVESTIGATION BY CLUSTERING PACKAGE Roman Melnyk 1), Yuriy Roman 2)
1) Software department, Lviv Polytechnic National University, The paper contains research results of graphic file format influence on image structure, quantitative changes in structuring coefficients of clustered images, influence of deformation changes caused by destruction processes on image quality. Clustering package for visual black-white and grey patterns was taken as the main instrument for investigation.
ESTIMATION OF TRANSMISSION RATE DEPENDENCE IN WI-FI Bogdan Gumenyuk 1), Grygoriy Vaskiv 2), Volodymyr Yankevych 3)
1) Lviv Polytechnic National University, This paper shows the results of research in to data transmission rate in Wi-Fi networks from influence the topology of the network, information character, services of enciphering and authentification.
THE APPROACH TO OPTIMAL DISTRIBUTION OF TASKS IN THE Mary Yurich
Zaporizhzhya National Technical University, In given article the approach to statement of a problem of optimal distribution of tasks in the computing system consisting of computers of various capacity and streams of tasks with different total duration various, leaning on the basic mathematical positions of a classical transport problem is considered, but considering thus specificity of a problem that will allow to find further more effective decision for such problems. It is proved, that the problem of distribution of tasks in the computing system can be considered as the modified transport problem. EFFICIENT PREPROCESSING FOR WEB LOG COMPRESSION Sebastian Deorowicz 1), Szymon Grabowski 2)
1) Politechnika Śląska, Instytut Informatyki, Web log files, storing user activity on a server, may grow at the pace of hundreds of megabytes a day, or even more, on popular sites. They are usually archived, as it enables further analysis, e.g., for detecting attacks or other server abuse patterns. In this work we present a specialized lossless Apache web log preprocessor and test it with combination of several popular general-purpose compressors. Our method works on individual fields of log data (each storing such information like the client’s IP, date/time, requested file or query, download size in bytes, etc.), and utilizes such compression techniques like finding and extracting common prefixes and suffixes, dictionary-based phrase sequence substitution, move-to-front coding, and more. The test results show the proposed transform improves the average compression ratios 2.70 times in case of gzip and 1.86 times in case of bzip2.
ANALOG-DIGITAL CONVERSION MONTE-CARLO IN FACILITIES OF Lavriv M.V. 1), Petryshyn L.B. 2)
1) Precarpathian National University by V. Stephanyk, department of informatics, Application of method of analog-digital conversion of Monte Carlo is grounded as such, that allows considerably to extend the spectral bar of signal of conversion n in the class of integrating converters. Application of methods of generation of pseudorandom signals is offered on the basis of recursive and with the mirror mutual reflection of binary digits generators. High-quality descriptions and type of distributing are definite as even. Methods are developed and the structures of analog-digital converters are resulted, and also sentinel diagrams of their functioning. Directions and applied tasks of effective application of method of Monte Carlo are definite.
CLASSIFICATION OF METHODS OF THE SIGNAL SEMANTIC Marina Polyakova 1), Victor Krylov 2)
1) Applied Mathematic Department, Odessa National Polytechnic University, In this paper the methods of signal semantic wavelet transform which underlines the edges of image in the edge detector task are analyzed and classified.
NEURAL-NETWORK BASED METHOD OF CONTROL Iryna Turchenko
Research Institute of Intelligent Computer Systems A simulation model of a section of mine ventilation network is considered in this paper. The simulation modeling of transient aerogasdynamic processes of methane concentration changing is fulfilled at applying position and exponential control influences. There is proposed a neural-based method of control influences forming by neural network training on the set of optimal control influences. There are defined a criterion and developed an algorithm of optimal control influences forming as a training set of neural network. The simulation modeling of applying of control influences formed by neural network is fulfilled and decreasing of control parameter in the section of mine ventilation network is estimated. FUZZY KNOWLEDGEBASE DESIGN FOR INTELLECTUAL SYSTEMS Larisa S. Globa 1), Maksym Y. Ternovoy 2), Olena S. Shtogryna 3)
1) National Technical University of Ukraine “Kiev Polytechnic Institute”, The approach to fuzzy knowledgebase designing for complex administrative intellectual systems is proposed in the paper. It consists of two phases: structure construction and membership function parameters adjustment. This approach allows accelerate the intellectual systems fuzzy knowledgebase construction.
KNOWLEDGE-BASED SIGNAL PROCESSING Jan Matuszewski
Institute of Radioelectronics, Faculty of Electronics, The paper presents some aspects of radar signals acquisition in the electronic intelligence (ELINT) system, the analysis of their parameters, feature extraction using linear Karhunen-Loeve transformation and applying knowledge-based techniques to the recognition of the intercepted signals. The process of final emitter identification is based on “the knowledge-based approach” which was implemented during the processing of constructing the database.
OPTOELECTRONNICS FUNCTION AND ALGORITHM-SPECIFIC INFORMATIVE Igor Lytvyn 1), Valeriy Martynyuk 2)
1) Department of Automated Systems and Programming, Presented analysis of influence of parameters and descriptions on informative processes passing in Optoelectronics Function and Algorithm-Specific Informative Devices units for providing a rational choice of information flow diagrams and hardware for their realization under the concrete task.
DIGITAL SIGNATURE PROTOCOLS Anna Nelasa 1), Victor Dolgov 2), Anatolij Pogorily 3)
1) Computer's Department, Zaporizhzhya National Technical University, In this paper the modified algorithms of the individual and collective digital signature on elliptic curves with fast verification for a corporate network are offered. BIOMEDICAL IMAGE SEARCH AND RETRIEVAL ALGORITHMS O. Berezsky 1), G. Melnyk 2), Yu. Batko 3)
1) Ternopil National Economic University, In this paper algorithm for search the tumour cells images in a database is developed. This algorithm based on shape and colour image features.
HUMAN FACE DETECTION METHODS USING Ihor Paliy
Research Institute of Intelligent Computer Systems The paper presents the improved human face detection method using the combined cascade of classifiers with the improved face candidates’ verification approach, as well as methods and algorithms for the verification level (convolutional neural network) structure generation and training. The combined cascade shows a high detection rate with a very small number of false positives and the proposed candidates’ verification approach is in almost 3 times faster in comparison with the classic verification scheme. The network’s structure generation method allows creating the sparse asymmetric structure of the convolutional neural network automatically. The improved training method uses the adaptive training examples ratio to obtain a trained network with a very low classification error for the positive examples.
DEVELOPMENT RESULTS OF WIRED SPECIALIZED DIGITAL Shokalo V.Ì., Tsopa O.I., Maslyi V.V.
Kharkiv National University of Radio Electronics, Conception of special digital information transmission systems (SDITS) creation and results of decision of Ukraine actual scientifically-applied problem of domestic connection apparatus development is presented on the basis of xDSL technologies for the existent wired communication channels. The example of specialized information system development for UMIA is presented in thesis.
LOGIC-TEMPORAL FUNCTIONS INTEGRATION V.P. Kozhemiako, N.V. Sachaniuk-Kavets’ka, L.O. Volontyr
Vinnytsia National Technical University In paper there are discussed problems of integration operation determination and definition of primary logical-time functions, characteristic of binary images integration for eye-processor images processing efficiency increasing and possibility of analog signal transformation into discrete numerical expression.
IDENTIFICATION OF MONOD-IYERUSALIMSKII SYSTEM Pasichnyk R. M. 1), Pigovsky Y. R. 2)
1) Economic Cybernetics Department, Ternopil National Economic University, Methods for structural and parametric identification of Monod-Iyerusalimskii system with loopback control are researched.
PROGRAM MODELS FOR INTERACTIVE DESIGN OF FAULT-TOLERANT Bohdan Mandziy, Bohdan Volochiy, Leonid Ozirkovskyy
Department of Theoretical Radio Engineering and Radiomeasurement The problem of development of program models of the fault-tolerant system with the mixed structural redundancy is solved. For designing of program models an advanced technology of analytical modeling of the systems is used. This technology takes into account all features of structure and behavior of the system under the action of refusals and two variants of realization of maintenance service strategy. It provides possibility of multiple analysis at the small expenses of time. The results of researches of the fault-tolerant system, which illustrate possibilities of decision of tasks of reliability analysis and reliability synthesis, are resulted in the article.
METHODS AND ALGORITHM FOR COLOR METALLOGRAPHIC Kornij V.V.
Physics and Mechanical Institute, The problem of processing of coloured 3D metallographic images in order to obtain the crack volume is considered. Method of 3D surface reconstruction is applied for crack volume estimation. The model of illumination image forming and algorithm of norm vector for horizontal cracks definition are developed. Formulas for crack volume estimation in the case of 3D reconstruction are presented.
THE USE OF NEURAL NETWORKS IN RARE VEGETATION E.M. Gambarova
Institute for Aerospace Informatics, This paper describes training of Multilayer Perceptron Neural classifier to extract rare vegetation objects from high spatial resolution IKONOS satellite imagery. There have been considered three options of training of the Multilayer Perceptron Neural according to three different classification schemes. At first 12 type of rare vegetation community types were defined, a main classification scheme (“Initial classification scheme”) was designed on that base. After prelim statistical tests on training samples two modification algorithms of the classification scheme were defined: the first one led to creating of scheme consisting of 7 classes (“Modified classification scheme”) and second one led us to creating of 5-classes scheme (“Optimized classification scheme“). The learning procedures of these classifiers are described as well as analysis and post processing of extraction results of objects of interest using Geoinformation Technologies in details.
THE TERNOPIL EDUCATIONAL COMMUNICATION CENTER:
George Markowsky 1), Anatoly Sachenko 2), Serhiy Voznyak 2), Viktor Spilchuk 2),
1) Department of Computer Science, University of Maine The Ternopil Educational Communication Center was developed with NATO funding to improve computing for the universities of Ternopil, Ukraine. It provides high speed access for all of Ternopil’s universities to the Internet and the World Wide Web. It enables high speed communication within Ternopil, and between institutions in Ternopil using the UarNet and URAN networks. It provides an integration of on-line library services, and it supports such activities as video conferencing, distance education, and class to class participation. The universities of Ternopil and the University of Maine are developing programs and projects to involve students and faculty both in Ukraine and the US. |