关键词: 关联规则, 免疫克隆算法, 文化算法, 自适应变异算子, 双层进化机制

Abstract: For the association rules mining, a method of mining association rules based on cultured immune clone algorithm is proposed. This method uses two-layer evolutionary mechanism and embeds the immune clone algorithm in the culture algorithm framework. It uses the intelligent searching ability of the immune clone algorithm and the commonly accepted knowledge in the culture algorithm to guide the rules mining. The situational knowledge and history knowledge in the culture algorithm are redefined, and a new mutation operator is put forward. This operator has the adaptive adjustment of mutation measure to improve the global search ability of immune clone algorithm. The experiments show that the new algorithm is superior to immune clone algorithm in performance speed and the rules’ accuracy.

Key words: association rules, immune clone algorithm, culture algorithm, self-adaptive mutation operator, two-layer evolutionary mechanism