Common sense knowledge based hybrid interestingness measures for data mining

Abstract

The association rule mining is now widely used in many fields such as commerce, telecom, insurance, and bioinformatics. Though it is improved in performance, the real commerce database size and dimension has greatly increased to a point of creating thousands or millions of association rules. In spite of using minimum support and confidence thresholds to help weed out or exclude the exploration of uninteresting rules, many rules that are not interesting to the user may still be produced. We develop intelligent data mining technique that generate and evaluate association rules by hybrid interestingness measures based common sense knowledge. We provide new and interesting knowledge to users by Common-Sense Measures. We define a Common-Sense Measures by similarity between association rules and common sense knowledge. This measure is based on the common sense knowledge network. © 2012 Springer-Verlag.