A Method for Finding Implicating Rules Based on the Genetic Algorithm

In information system, some rules have implicating relations (called implicating rules), but some rules have not. An approach of finding implicating rules based on the genetic algorithm is proposed. Some properties of independence and correlation of descriptions are discussed. It can obtain directly the implicating rules (including the positive and negative rules) with the correlation of two descriptions according to strength of implication defined in this paper. At the same time, an algorithm for finding optimized rules based on genetic algorithm is presented. By it, the problem of efficiency of finding rules is solved. At last, the efficiency and practicability of the method are illustrated by the experiment results.