Semantic network based common sense measure for association rule pruning

Ingi Lee, Hwan Seung Yong

Research output: Contribution to journalArticlepeer-review


Association rule mining is now widely used in many fields such as commerce, telecom, insurance, and bioinformatics. Although association rule mining has improved in performance, the real commerce database has also grown in its size and dimension to a point of creating millions of association rules. One of the biggest problems of association rule mining is that it frequently produces large numbers of rules, and this makes it difficult for users to select those that are of interest. We proposed the Common Sense Measure (CSM) so that only interesting knowledge can be selected in order to resolve the problem resulting from a large quantity of rules. The CSM is an interestingness measure that evaluates how closely rules match the common sense knowledge. We developed an algorithm of rule matching method with common sense knowledge using the common sense network (CSN).

Original languageEnglish
Pages (from-to)183-192
Number of pages10
JournalInternational Journal of Bio-Science and Bio-Technology
Issue number4
StatePublished - 2013


  • Common sense knowledge
  • Data mining
  • Interestingness measures
  • Knowledge representation
  • Semantic network
  • Similarity


Dive into the research topics of 'Semantic network based common sense measure for association rule pruning'. Together they form a unique fingerprint.

Cite this