Mining Actionable Behavioral Rules Based on Decision Tree Classifier
Peng Su, Jian Yang, Zhenpeng Li, Yuan Liu · 2017
Actionable behavioral rules mining is a new problem of data mining. The produced rules can provide the user explicit suggestions of actions to influence the behaviors of the entity in concern with satisfactory utility to the user. To guarantee the reliability of the rules, the traditional mining approaches need to find frequent action sets. However, this will result in high time complexity. In this paper, to handle this problem, a decision-tree-classifier-based mining algorithm are proposed. It achieves reduction of time complexity by avoiding finding frequent action sets. The experimental results strongly suggest the superiority of our approach.