Abstract: Decision making processes in manufacturing companies are becoming extremely complex and require more and more knowledge, both of technological quality of products, concerning a production process, as well as the industrial engineering and control. The increase in the scale of production and the level of technological development have caused that industrial companies have become systems that require the application of modern effective methods of the decision making. This paper presents a study of the issues related to decision-making and knowledge acquisition in enterprises. The model of intelligent decision support system for the production company has been developed. The paper presents the main idea of system and data models in the form of ER diagram, that will finally be implemented in RDBMS environment. In addition, an attempt to assess such systems.

Keywords: decision support systems, knowledge base, knowledge representation, knowledge acquisition.

ACM Classification Keywords: I. Computing Methodologies, I.2.1 Applications and Expert Systems, J. Computer Applications,

Link:

THE MODEL OF DECISION SUPPORT SYSTEM FOR A MANUFACTURING COMPANY

Monika Piróg-Mazur?

http://foibg.com/ibs_isc/ibs-27/ibs-27-p09.pdf

Abstract: We consider the condition of the same ordering of objects due to using the convolution of utility functions and membership functions. It turns out that the same order takes place when the each utility function of an attribute is constructed as a convolution of fuzzy membership functions of this attribute. To transform membership functions to utility function of an attribute the formula was deduced. The example illustrates that transforming. The paper provides examples of application of the transformation of membership functions in the utility function, and vice versa.

Keywords: utility function, membership function, additive and multiplicative convolution.

ACM Classification Keywords: G. Mathematics of Computing, I.2.1 Applications and Expert Systems.

Link:

STUDY RELATIONSHIP BETWEEN UTILITY FUNCTION AND MEMBERSHIP FUNCTION IN THE PROBLEM OF OBJECT RANKING1

Stanislav V. Mikoni, Marina I. Garina

http://foibg.com/ibs_isc/ibs-27/ibs-27-p08.pdf

Abstract: In the paper different well-known market characteristics and indicators of financial accounting (net income, revenue, revenue growth, etc.) are considered. We propose new characteristics, which could be useful for company assessment. We also describe a classification technique based on Naive Bayes method to identify the most attractive companies.

Keywords: classification, Naive Bayes, financial market, stock’s returns.

ACM Classification Keywords: I.2 ARTIFICIAL INTELLIGENCE

Link:

STUDY OF INVESTMENT ATTRACTIVENESS OF RUSSIAN COMPANIES ON THE BASIS OF THEIR MARKET CHARACTERISTICS AND PERFORMANCE REPORTING

Dmitriy Bogachev

http://foibg.com/ibs_isc/ibs-27/ibs-27-p07.pdf

Abstract: Time series prognosis of economical indexes is one of the main problems of econometrics. In the paper we study possibility to give an interval prognosis of time series using the set of the best prognostic models. Speaking ‘model’ we mean a combined model of regression and auto-regression. Speaking ‘the best models’ we mean the ordered series of models constructed by the well-known Group Method of Data Handling (GMDH). The proposed simple approach consists in the following: a) one generates the fixed numbers of models on the basis of experimental data b) these models give correspondent prognoses c) the real value is supposed to belong to min-max interval the models provide. We shortly describe the software tool GMDH-Shell (GS) that implements GMDH and the results of experiments with GS. The experimental data are time series of the Gross Domestic Products (GDP) of 100 countries given on the period 1980-2000.

Keywords: GMDH, GMDH Shell, time series prognosis, gross domestic product

ACM Classification Keywords: I.2 Artificial Intelligence

Link:

TIME SERIES PROGNOSIS OF GDP WITH THE SYSTEM GMDH-SHELL (EXPERIMENTAL WORK)

Victor Lebedev

http://foibg.com/ibs_isc/ibs-27/ibs-27-p06.pdf

Abstract: Smoothing and prognosis belong to the main problems, which we deal with when process various time series. Unlike the typical approaches related with lineal regression on polynomial and trigonometric functions we consider local procedures. We describe algorithms that resolve these problems and present results of experiments using data of the Russian State Statistical Committee. All procedures are realized in the Excel-VBA.

Keywords: econometrics, local regression, time series, seasonality

ACM Classification Keywords: I.2.m Miscellaneous

Link:

SMOOTHING AND PROGNOSIS OF MULTI-FACTOR TIME SERIES OF ECONOMICAL DATA BY MEANS OF LOCAL PROCEDURES (REGRESSION AND CURVATURE EVALUATION)

Alexander Kovaldji, Vladimir Averkiev, Marina Sarkissyan

http://foibg.com/ibs_isc/ibs-27/ibs-27-p05.pdf

Abstract: The article is devoted to the issue of using financial multipliers to estimate the value of a company. In our research we identified the industries that account for more than 50% of Gross Domestic Product. We took the companies with publicly available information. Then we used multicriteria Muchnik's model to identify promising companies and receive the required sample. At the final stage, we used a comparative approach for the evaluation of companies in these industries. The results of experiments showed the essential advantage of the proposed method. Notably, we marked multiplier, which were most likely gave a more realistic evaluation of companies. The paper reflects the results of Bachelor research.

Keywords: financial multiples; company valuation; multicriteria Muchnik's model; comparable company

ACM Classification Keywords: I.2.m Miscellaneous

Link:

ESTIMATION OF THE INVESTMENT ATTRACTIVENESS OF COMPANIES USING MULTIPLES THAT ACCOUNT FOR INDUSTRY SPECIFIC FACTORS

Evgeniy Ageev

http://foibg.com/ibs_isc/ibs-27/ibs-27-p04.pdf

Abstract: Forrester model is a system of differential equations reflecting dynamics of 5 macro-economical variables (population, resources, etc.) . Such a model were developed more then 40 year ago and it proved to be an effective tool for qualitative analysis of world dynamics. By the moment there is no an accessible end-user program based on this model and our goal was the development of such a program with a comfortable graphical interface. In the paper we describe the program World-Dyn?, which allows to set initial data and noise level, to set moments of parameter changes, to form the necessary visualization of results. We demonstrate the program functionality both on Forrester’s example and on our example related with crisis.

Keywords: Forrester model, world dynamics, numerical analysis.

Link:

PROGRAM WORLD-DYN BASED ON FORRESTER MODEL OF WORLD DYNAMICS

Olga Proncheva

http://foibg.com/ibs_isc/ibs-27/ibs-27-p03.pdf

Abstract: Decision trees are the well-known a popular tool in the areas where non-numerical data are used. In the paper we demonstrate the application of decision trees for analysis of International logistics. First of all we cluster data to reduce the number of variants for comparison. Then the regression analysis is completed and the significant parameters combinations are determined. The logical function and decision tree could be useful for revealing possible corruption schemes in logistics. Finally we complete experiments with this decision tree. The suggested method can be used for constructing a decision tree based on a large amount of similar data.

ACM Classification Keywords: 1.2 Artificial Intelligence

Keywords: decision trees, logistics, corruption

Link:

CONSTRUCTING DECISION TREE BASED ON QUESTIONNAIRES TO DETECT A POSSIBLE CORRUPTION IN LOGISTICS

Dmitry Stefanovskiy, Sergey Maruev, Xavier Tejada

http://foibg.com/ibs_isc/ibs-27/ibs-27-p02.pdf

Abstract: We consider queuing systems as models of distance learning system we analyze how characteristics of business process in the system affect on throughput and on learning outcomes. The processes of executing tasks and tasks validation process are the key processes in the distance learning. A model of the process performed by a student is a queuing system with refusals. A model of the process performed by the teacher is a multi-channel queuing system with limited queue. We present a structure of one of the courses of the University, where the authors work to form individual trajectory of learning. for students with different levels of knowledge. Such an approach allow to increase the throughput of distance learning system.

Keywords: distance learning, process modeling, queuing system analysis

ACM Classification Keywords: K.3.1 Computer Uses in Education, Distance Learning

Link:

ANALYSIS OF THE THROUGHPUT OF THE PROCESS OF DISTANCE LEARNING

Sergey Maruev, Evgeniya Gorbunova

http://foibg.com/ibs_isc/ibs-27/ibs-27-p01.pdf

Abstract: This paper describes the principles for the development of interactive query or question-answer expert systems (ES) in the Multi Studio software environment in Multi (the universal language), as well as in such environments as Visual Prolog and CLIPS. The paper also presents a comparative analysis of those principles checked by solving the test problems to show a significant advantage of Multi in the development of those ES according to the main quality parameters: language usability, work content of ES development, electronic memory size and speed of the developed ES. As a result of our study we formulate several principles of developing ES tool environments:  Development of a super-high-level universal environment language by integrating the best qualities of all the languages considered in the paper.  Development of a universal structure of a knowledge base with a unitized syntax based on semantic networks.  Development of a hybrid tool environment which can separately perform the following:  A one-time translation (compilation) of program and data input texts into an internal language of a knowledge base  Multiple fetch of programs from a knowledge base (the programs are executed by interpretation of those programs’ instructions in a hybrid environment)

Keywords: expert systems, CLIPS, Multi Studio, Prolog.

ACM Classification Keywords: I.2.5 Programming Language and Software.

Link:

PRINCIPLES OF THE DEVELOPMENT OF INTERACTIVE QUERY EXPERT SYSTEMS

Valentin Kataev

http://www.foibg.com/ibs_isc/ibs-26/ibs-26-p24.pdf

Abstract: In the multi-attribute utility theory the utility functions are usually constructed by dots. It concerns both the lottery’s method and the value increasing method. In the both cases the utility function is constructed in the absolute scale 0,1 that causes inconveniences for experts. The comparative assessments look more preferable for decision-makers. The paired comparison matrix (PCM) looks as a natural model representing the preference structure of decision-maker (DM). We use scale points of attributes as a PCM comparative entities. We use also increasing/decreasing entity priority as a criterion. Function of priorities is transformed to utility function on the base of a normalizing function. Such a function allows using the matrix power as parameter affecting the form of utility function. The PCM provides the extended possibilities to DMs to form comparative assessments both the qualitative ones (as better-worse) and the quantitative ones reflecting winnings and losses of DMs. In the paper we consider methods for utility function construction having different forms of its presentation. Among them there are utility functions based on attributes measured in nominal scales.

Keywords: utility function, paired comparison matrix, scale points, priority function.

ACM Classification Keywords: H.4.2 Information Systems Applications: Types of Systems-decision support

Link:

UTILITY FUNCTION DESIGN ON THE BASE OF THE PAIRED COMPARISON MATRIX*

Stanislav Mikoni

http://www.foibg.com/ibs_isc/ibs-26/ibs-26-p23.pdf

Abstract: This paper presents an exploratory study of the effectiveness of support vector machines used as a tool for computer-aided breast cancer diagnosis. We explore the discriminatory power of heterogeneous mammographic and sonographic descriptors in solving the classification task. Various feature selection techniques were tested to find a set of descriptors that outperforms those from similar studies. We also explored how choice of the SVM kernel function and model parameters affect its predictive abilities. The kernels explored were linear, radial basis function, polynomial, and sigmoid. The model performance was estimated by ROC analysis and metrics, such as true and false positive rates, maximum accuracy, area under the ROC curve, partial area under the ROC curve with sensitivity above 90%, and specificity at 98% sensitivity. Particular attention was paid to the latter two as lack of specificity causes unnecessary surgical biopsies. Experiments registered that an appropriate reduction of variables can greatly improve the predictive power of the model, as long as the choice of the kernel affects the model performance marginally. We also found that the SVM is superior to the common classification technique used in the field - MLP neural networks.

Keywords: data mining, support vector machines, heterogeneous data; breast cancer diagnosis, computer aided diagnosis.

ACM Classification Keywords: I.5.2- Computing Methodologies - Pattern Recognition – Design Methodology - Classifier design and evaluation.

Link:

SUPPORT VECTOR MACHINES FOR CLASSIFICATION OF MALIGNANT AND BENIGN LESIONS

Anatoli Nachev, Mairead Hogan

http://www.foibg.com/ibs_isc/ibs-26/ibs-26-p22.pdf

Abstract: The use of Machine Learning (ML) techniques is already widespread in Medicine Diagnosis. The use of these techniques helps increasing the efficiency of human diagnostic, which is significantly affected by the human conditions such as stress as well as the lack of experience. In this paper, integration between two ML techniques case-based reasoning (CBR) and artificial neural network (ANN) is used for the automation of the diagnosis of pathology on the vertebral column. CBR is used for indexing and retrieval. For adaptation, an untrained ANN is fed with the retrieved closest matches. Then the ANN is trained and queried with the new problem to give the adapted solution. Experiments are conducted on the vertebral column data set from University of California Irvine (UCI) machine learning repository. A comparison with several machine learning techniques used for classifying the same problem is performed. Results show that the hybridization between CBR and ANN helps in improving the classification.

Keywords: Computer Aided Diagnosis System, Hybrid Intelligent Classifier, Vertebral Column, Case-Based? Reasoning, Artificial Neural Network.

ACM Classification Keywords: I.2.5 Expert system tools and techniques - Conference proceedings.

Link:

A HYBRID INTELLIGENT CLASSIFIER FOR THE DIAGNOSIS OF PATHOLOGY ON THE VERTEBRAL COLUM

Essam Abdrabou

http://www.foibg.com/ibs_isc/ibs-26/ibs-26-p21.pdf

Abstract: The clustering problem for multivariate observations often encountered in many applications connected with Data Mining and Exploratory Data Analysis. Conventional approach to solving these problems requires that each observation may belong to only one cluster, although a more natural situation is when the vector of features with different levels of probabilities or possibilities can belong to several classes. This situation is subject of consideration of fuzzy cluster analysis, intensively developing today. In many practical tasks of Data Mining, including clustering, data sets may contain gaps, information in which, for whatever reasons, is missing. More effective in this situation are approaches based on the mathematical apparatus of Computational Intelligence and first of all artificial neural networks and different modifications of classical fuzzy c-means (FCM) method. But these methods are effective only in cases when the original data set is given beforehand and does not change during data processing. At the same time there is a wide class of problems when the data are fed to processing sequentially in on-line mode as it occurs in self-organizing Kohonen networks training. At the same time apriori it is not known which of the vectors-images contain gaps. In this paper the problem of probabilistic and possibilistic on-line clustering of data with gaps using Partial Distance Strategy is discussed and solved, self-organizing neuro-fuzzy Kohonen network and new self-learning algorithm that is the hybrid of "Winner-takes-more" rule and recurrent fuzzy clustering procedures are proposed and investigated.

Keywords: Fuzzy clustering, Kohonen self-organizing network, learning rule, incomplete data with gaps.

ACM Classification Keywords: 1.2.6 Artificial Intelligence: Learning – Connectionism and neural nets; 1.2.8 Artificial Intelligence: Problem Solving, Control Methods, and Search – Control theory; 1.5.1 Pattern Recognition: Clustering – Algorithms.

Link:

ADAPTIVE CLUSTERING OF INCOMPLETE DATA USING NEURO-FUZZY KOHONEN NETWORK

Yevgeniy Bodyanskiy, Alina Shafronenko, Valentyna Volkova

http://www.foibg.com/ibs_isc/ibs-26/ibs-26-p20.pdf

Abstract: The paper presents the experiment on the time series whose elements are month values of BIS effective exchange rate of USD from January 1994 till March 2010. A tendency of BIS (Bank of International Settlements) effective exchange rate to increase or decrease is an expected value. First, a process of building of the neural network for events forecasting is presented, that is the selection of networks’ architecture and parameters. Next, the effect of adding data calibrated by nonlinear input function to input data calibrated linearly is described. The nonlinear input function - hyperbolic tangent was accepted. Hyperbolic tangent sigmoid transfer function and log sigmoid transfer function are commonly used as transfer functions in neural networks.

Keywords: neural network, time series.

ACM Classification Keywords: I.2.8 Data calibration.

Link:

THE EFFECT OF INTRODUCTION OF THE NON-LINEAR CALIBRATION FUNCTION AT THE INPUT OF THE NEURAL NETWORK

Piotr Romanowski

http://www.foibg.com/ibs_isc/ibs-26/ibs-26-p19.pdf

Abstract: The main goal of this paper is to present and evaluate the possibility of using the methods and tools of Artificial Intelligence and Data Mining to analyze marketing data needed to support decision-making in the process of market segmentation. This paper describes the application of Kohonen’s Neural Networks and Classification Trees (including tools such as CART-Classification and Regression Tree, Chi-squared Automatic Interaction Detector (CHAID) and Boosted Tree) to solving problems of classification and grouping of data. The main part presents the results of market segmentation that can be used by the company producing household products. Finally conclusions and further research plans have been described.

Keywords: data analysis, artificial intelligence, data mining, classification, clustering, Kohonen’s neural networks.

ACM Classification Keywords: I.2.m Miscellaneous : I.2.6 Artificial Intelligence: Learning – Connectionism and neural nets; I.5.1: Models – Neural nets; I.5.3: Clustering – Algorithms.

Link:

INTELLIGENT ANALYSIS OF MARKETING DATA

Łukasz Paśko, Galina Setlak

http://www.foibg.com/ibs_isc/ibs-26/ibs-26-p18.pdf

Abstract: The Effective methods of intelligent analysis of business processes, in particular, methods of revealing fragments of such processes are developed. Besides, analyzing information extracted from journals of registering events of a business process (BP) to formalize the real behavior of a BP is carried out. Such data analysis is especially important in those cases when the occurring sequence of events is registered, i.e. executives have an opportunity to make a decision about the order of further process implementation.

Keywords: business process, procedure, logical net, intelligent analysis.

ACM Classification Keywords: I.2 Artificial Intelligence – Knowledge Representation Formalisms and Methods

Link:

INTELLIGENT METHODS OF REVEALING FRAGMENTS IN BUSINESS PROCESSES

Nataliia Golian, Vira Golian, Olga Kalynychenko

http://www.foibg.com/ibs_isc/ibs-26/ibs-26-p17.pdf

Abstract. The main goal of BI systems is to provide the access for the users to crucial information connected with the tools they use every day. It allows to take more relevant decisions, share knowledge with other people, cooperate within the whole organization and increase the company's gainings. The offered functionality includes either the scalable technology's platforms designed for workers in all management tiers.

Keywords: Business Intel's lIntelligence, Business Discovery, information, analysis, Qlickview

ACM Classification Keywords: K.6 Management of computing and information systems - K.6.0 General economics

Link:

BI – SUPPORTING THE PROCESSES OF THE ORGANIZATION'S KNOWLEDGE MANAGEMENT

Justyna Stasieńko

http://www.foibg.com/ibs_isc/ibs-26/ibs-26-p16.pdf

Abstract: Inductive modelling tools are widely used for solving problems of analysing economical, ecological, and other processes. Development of business intelligence systems based on inductive modelling algorithms for analysis, modelling, forecasting, classification, and clustering of complex processes is very promising. When solving real tasks of model construction from statistical data, the question of storage of and providing effective access to the information arises. At the stage of input data processing there are typical difficulties with processing data in different formats as well as containing omissions and untypically small or big values etc. From the other side, the question of output information storage exists like determination of structure and parameters of models, estimation of precision and validity, plots and diagrams drawing etc. This would allow structuring input data of different types and using the information already existing in database and also provide the storage of complete information on experiments and results of calculations. To solve such kind of problems, the integrated environment for storing and handling information is developed. Architecture of the environment is offered giving the possibilities to manipulate present information freely using relational database containing only metadata and storing input statistical data and output results of calculations.

Keywords: integrated environment, handling and storing information, inductive modeling, GMDH-algorithms, Business Intelligence

ACM Classification Keywords: H.2.8 Data Base Application – Data Mining

Link:

INTEGRATED ENVIRONMENT FOR STORING AND HANDLING INFORMATION IN TASKS OF INDUCTIVE MODELLING FOR BUSINESS INTELLIGENCE SYSTEMS

Nataliya Shcherbakova, Volodymyr Stepashko

http://www.foibg.com/ibs_isc/ibs-26/ibs-26-p15.pdf

Abstract: The classical Forrester model of world dynamics is a system of 5 differential equations related with 5 macro-economical variables (population, resources, etc.). This model was developed at 1970-1971 but by the moment its stability to noise was not studied. The plan of experiments is described and the results of modeling are presented. It proved that a) noise affects stronger initial data then the model during its functionality b) change of resources is the most critical value in comparison with the other system variables. All experiments have been made by means of the program WorldDyn? developed on MatLab?.

Keywords: Forrester model, word dynamics noise immunity, numerical analysis

Link:

TESTING STABILITY OF THE CLASSICAL FORRESTER MODEL TO INITIAL DATA AND ADDITIVE NOISE

Olga Proncheva, Mikhail Alexandrov, Sergey Makhov

http://www.foibg.com/ibs_isc/ibs-26/ibs-26-p14.pdf

Abstract: At far 1970 the elite Roman Club asked prof. J. Forrester from MIT to develop a model of world dynamics. Speaking world dynamics we mean the dynamic interactivity of the main macro economical variables. The 1-st version of the model named “World-1” was presented in 4 weeks and next year the corrected version “World-2” was accepted as the classical J. Forrester’s model. In spite of its long history the J. Forrester model retains its actuality being the basis for modern models. In the paper we consider the principal of system dynamic, criticism of the classical model, and the new models developed by the J. Forrester’s followers. We consider also adjacent areas and open problems related with world dynamics.

Keywords: world dynamics, non lineal dynamics, J. Forrester model

ACM Classification Keywords: I.2.m Miscellaneous

Link:

J. FORRESTER’S MODEL OF WORLD DYNAMICS AND ITS DEVELOPMENT (REVIEW)

Olga Proncheva, Sergey Makhov

http://www.foibg.com/ibs_isc/ibs-26/ibs-26-p13.pdf

Abstract: We investigate a possible way for solving the problem of combination of logical inference search methods and symbolic computation tools in e-learning testing on the basis of the approaches developed at the Kiev schools of automated theorem proving and analytical transformations. The investigations started in the first half of 1960s at the Institute of Cybernetics of the Academy of Sciences of Ukraine. Some years later the Faculty of Cybernetics of the Kiev State University was involved in the corresponding projects. The current state of investigations on the topic as well as their theoretical and practical background is described in the paper.

Keywords: analytical transformation, automated theorem proving, deduction, e-learning, intelligent tutoring system.

ACM Classification Keywords: I.2.3 Deduction and Theorem Proving – Deduction. I.2.4 Knowledge Representation Formalisms and Methods – Predicate logic. G.4 Mathematical software. K.3.2 Computer and Information Science Education.

Link:

ON COMBINATION OF DEDUCTION AND ANALYTICAL TRANSFORMATIONS IN E-LEARNING TESTING

Vitaly Klimenko, Alexander Lyaletski, Mykola Nikitchenko

http://www.foibg.com/ibs_isc/ibs-26/ibs-26-p12.pdf

Abstract: In the paper global modeling of complex systems with regard to quality of local models of simple plants is discussed. Complex systems consists of several sub-systems. As a global model multilayer feedforward neural networks are used. It is desirable to obtain an optimal global model, as well as optimal local models. A synthetic quality criterion as a sum of the global quality criterion and local quality criteria is defined. By optimization of the synthetic quality criterion can be obtained the global model with regard to the quality of local models of simple plants. The quality criterion of the global model contains coefficients which define the participation of the local quality criteria in the synthetic quality criterion. The investigation of influence of these coefficients on the quality of the global model of the complex static system is discussed. The investigation is examined by a complex system which is composed from two nonlinear simple plants. In this paper complex system means real chemical object (i.e. a part of the line production of ammonium nitrite).

Keywords: complex system, neural network, global modeling

ACM Classification Keywords: I.2.6 ARTIFICIAL INTELLIGENCE, Learning - Connectionism and neural nets

Link:

STUDY THE QUALITY OF GLOBAL NEURAL MODEL WITH REGARD TO LOCAL MODELS OF CHEMICAL COMPLEX SYSTEM

Grzegorz Drałus

http://www.foibg.com/ibs_isc/ibs-26/ibs-26-p11.pdf

Abstract: Decomposition methods for solving large-scale Traveling Salesman Problem (TSP) are presented. Three approaches are proposed: macromodeling for clustered TSP as well as extending and “ring” methods for arbitrary points’ distribution. Four stages of the problem solving include partitioning of the input set of points into small subsets, finding the partial high quality solutions in the subsets, merging the partial solutions into the complete initial solution and optimizing the final solution. Experimental investigations as well as the comparative analysis of the results and their effectiveness estimation in terms of quality and running time were conducted. The suggested approaches provide substantial reduction in the running time in comparison with the existing heuristic algorithms. The quality loss is small. The problem instances up to 200,000 points were investigated. The TSP is extensively applied in transportation systems analysis, printed circuit boards, VLSI, SoC and NoC computer-aided design, testing and manufacturing, laser cutting of plastics and metals, protein structure research, continuous line drawings, X-ray crystallography as well as in number of other fields.

Keywords: traveling salesman problem, combinatorial NP-hard problems, decomposition, large-scale.

ACM Classification Keywords: G.2.1 Combinatorics - Combinatorial algorithms; I.2.8 Problem Solving, Control Methods, and Search - Heuristic methods.

Link:

DECOMPOSITION METHODS FOR LARGE-SCALE TSP

Roman Bazylevych, Marek Pałasiński, Roman Kutelmakh, Bohdan Kuz, Lubov Bazylevych

http://www.foibg.com/ibs_isc/ibs-26/ibs-26-p10.pdf

Abstract: The aim of the paper is to present the assumptions and the architecture of the system for searching similarity in string sets. During the research all the required steps of a procedure of text documents processing which includes text extraction, pruning, stemming and lemmatization were analysed. Models of a text documents’ description and the method of creating a vector of features were developed as well. This vector consists, inter alia, of chosen words and the number of their occurrences. The process of the text analysis is supported by a set of various dictionaries. These are Stop-words, Domain and Lemma dictionaries and all of them were considered in the context of the Polish language. Because the Lemma dictionary is supposed to consist of many entries, the efficient method of its access optimisation was elaborated. Various measures used for calculating degree of a text documents similarity were studied too. Moreover, the method for determining the quality of user queries and text documents adjustment were proposed. The system was realized in accordance with the idea of multi-agent systems. Its functionality is ensured by the set of agents acting on the basis of separate threads. In the research, tests of the system work efficiency were also performed.

Keywords: agent systems, text similarity search

ACM Classification Keywords: I.7 Document And Text Processing

Link:

MULTI-AGENT SYSTEM FOR SIMILARITY SEARCH IN STRING SETS

Katarzyna Harężlak, Michał Sala

http://www.foibg.com/ibs_isc/ibs-26/ibs-26-p09.pdf