Genetic network programming with acquisition mechanisms of association rules in dense database

A method of association rule mining using Genetic Network Programming (GNP) is proposed to improve the performance of association rule extraction from dense database. Rule extraction is done without identifying frequent itemsets used in Apriori-like methods. Association rules are represented as the connections of nodes in GNP. The proposed mechanisms calculate measurements of association rules directly from a database using GNP, and measure the significance of the association via the chi-squared test. The proposed system evolves itself by an evolutionary method and obtains candidates of association rules by genetic operations. Extracted association rules are stored in a pool all together through generations and reflected in genetic operators as acquired information. In this paper, we describe an algorithm capable of finding important association rules using GNP with sophisticated rule acquisition mechanisms and present some experimental results.