A Subgroup Discovery Algorithm Based on Genetic Fuzzy Systems

Abstract

Subgroup discovery algorithm is a new data mining technique, which plays an important role in the induction of large data areas. First, the basic concepts of subgroup discovery algorithm and fuzzy system are introduced. Then subgroup discovery iterative genetic algorithm (SDIGA) is studied. Genetic fuzzy system is used in traditional subgroup discovery algorithm, the way that a weighted sum of multiple objective functions is taken in fitness function. After continuous crossover genetic, the best description of the rules is obtained. Finally, the proposed method is applied to the dataset of compressive strength of concrete in UCI database, and the experiment results show the effectiveness of SDIGA subgroup discovery algorithm.