Research of the Subgroup Discovery Algorithm NMEEF-SD

摘要

Subgroup discovery (SD) is a data mining technique which could find the most interesting individual patterns from a population of individuals for the user. Non-dominated Multi-objective Evolutionary algorithm for Extracting Fuzzy rules in Subgroup Discover (NMEEF-SD) which is based on non-dominated sorting genetic algorithm Ⅱ (NSGA-Ⅱ) is a kind of algorithm for SD. First, the concept of subgroup discovery is introduced. Then NMEEF-SD algorithm and its main properties are researched. Finally, the algorithm is applied to analyze the concrete comprehensive strength dataset from UCI database, the result of experiment shows that the NMEEF-SD algorithm is able to extract fuzzy rules with interesting characteristics and is easy to understand.