Method for Discovering Actionable Rule
Hengmin Zhu, Weidong Huang, Hui-song Zheng · 2007
An important task for data mining is to discover interesting rules for the user. Interesting rules not only are novel to users, but also must be actionable to the user. A method to discover actionable rule is proposed in this paper. It adopts rule template to represent user's preference for actionable rule form. The concept of conformability measure of discovered rule relative to rule template is proposed. The factors which affect rule actionability are discussed and function to measure rule actionability is given. Experimental result shows that the method is effective.