关键词: 关联分类, 支持度阈值, 置信度阈值, Apriori_TFP, 爬山法

Abstract: The set of the support and confident thresholds usually affects the accuracy of classification based on association rules. As for the previous associative classification algorithms, the two thresholds are always set by experiences, so it is difficult to ensure that the classifier can always get the best accuracy. In order to solve this problem, the optimization strategies can be introduced to associative classification algorithm. The hill climbing search method was used to improve the Apriori_TFP_CMAR algorithm to get the highest classification accuracy of the set of support and confidence thresholds. This strategy can avoid the unreasonable set of threshold, and enhance the classification accuracy.

Key words: associative classification, support threshold, confident threshold, Apriori_TFP, hill climbing