Wróblewski J., 1998. A Parallel Algorithm for Knowledge Discovery System. Proc. of PARELEC'98, Bialystok, Poland. The Press Syndicate of the Technical University of Bialystok, Bialystok 1998, pp. 228 - 230.
ABSTRACT
A rough set based knowledge discovery system is presented. The system is based on the decomposition of large data tables into smaller ones in such a way that the approximation of global decision from local ones (related to these smaller tables) can be obtained. A set of locally generated rules can be collected for achieving the sufficient approximation of the global decision algorithm. A parallel (distributed) system for data analysis using local reduct-based rule generator is described and discussed. Experiment results on large data tables are presented.