AN OVERVIEW OF EXISTING MULTI-AGENT SYSTEMS FOR DATA MINING PROBLEM SOLVING

Abstract

The paper analyzes the multi-agent systems for data mining of large amounts of data, which are considered as a set of intellectual agents and relationships between them. It is shown that the process of data analysis includes the stage of data integration, pre-processing, classification, clustering, the definition of associative rules and the presentation of data in the form of a knowledge base. As evaluation parameters are used such parameters as: evaluation of work with data and the knowledge base creation, evaluation of system architecture organization and system processes execution, agents learning efficiency, evaluation of functional elements interaction, evaluation of “man – machine” interaction, efficiency of hardware and software resources usage. The example of a web-service indicates the problems that arise during a multi-agent system for data mining development process.

Authors and Affiliations

В. С. Яременко

Keywords

Related Articles

RAY CASTING OF VOLUME DATA AND FUNCTIONALLY SURFACED SURFACES FOR MEDICAL APPLICATIONS WITH APPLICATION OF GRAPHIC ACCELERATORS

Medical image three-dimensional reconstruction technology has been widely used in medical diagnosis and treatment planning. GPU based ray casting has become a valuable tool for the visualization of medical image data. Wh...

LAMINARIA IN THE PRODUCTION OF PROCESSED CHEESE

The prospects of developing a recipe for processed cheese with laminaria extract are proved. Problematic issues of iodine deficiency and ways to solve this problem are highlighted. On the basis of experimental studies an...

PURPOSE OF CPS EQUIPMENT. MODES OF OPERATION OF CPS FUNCTIONAL SYSTEMS. THE MODEL OF EFFICIENCY OF THE CPS REGULATORY GROUP AT ZAPORIZHIA NPP

One of the most important tasks in the operation of a nuclear reactor is to ensure nuclear safety and control its operating modes. The nuclear reactor is controlled mainly by the control and protection system (CPS), ther...

FLAKES OF HIGH NUTRITIONAL VALUE ARE ON THE BASIS OF THE BIOLOGICALLY ACTIVE GRAIN

The application of the regime of hydrothermal grain processing at temperature 12–16 oС is substantiated. Under these conditions activation of the enzyme complex, reduction of grain density and increase of its specific vo...

AUTOMATED QUALITY MANAGEMENT OF BAKERY PRODUCTION ON THE BASIS OF PRINCIPLES OF QUALIMETRY

The article explores methods of improving the quality of bakery products at all stages of production and certification for sale to consumers with the help of an integrated automated product quality management system that...

Download PDF file
  • EP ID EP579518
  • DOI -
  • Views 81
  • Downloads 0

How To Cite

В. С. Яременко (2018). AN OVERVIEW OF EXISTING MULTI-AGENT SYSTEMS FOR DATA MINING PROBLEM SOLVING. Вчені записки Таврійського національного університету імені В. І. Вернадського. Серія: Технічні науки, 29(3), 47-55. https://europub.co.uk/articles/-A-579518