Obtaining expert system rules using data mining tools from a power generation database

摘要

Four data mining tools were applied and evaluated on a real power generation database with thermoelectric and hydroelectric mexican utilities information from years 1988 to 1995. In this paper we present the results obtained using the tools C4.5, CN2, FOIL and PEBLS. We evaluated accuracy, knowledge amount reduction and processing time. Additionally, we describe the expert system rules (extracted knowledge) and we propose an architecture of an integrated knowledge discovery system for this power generation database.