A Bayesian scoring technique for mining predictive and non-spurious rules

Batal I, Cooper G, Hauskrecht M. A Bayesian scoring technique for mining predictive and non-spurious rules. Machine Learning and Knowledge Discovery in Databases 7524 (2012) 260-276.  PMID: 25938136 PMC4416489

Rule mining is an important class of data mining methods for discovering interesting patterns in data. The success of a rule mining method heavily depends on the evaluation function that is used to assess the quality of the rules. In this work, we propose a new rule evaluation score - the Predictive and Non-Spurious Rules (PNSR) score. This score relies on Bayesian inference to evaluate the quality of the rules and considers the structure of the rules to filter out spurious rules. We present an efficient algorithm for finding rules with high PNSR scores. The experiments demonstrate that our method is able to cover and explain the data with a much smaller rule set than existing methods.

Publication Year: 
2012
Publication Credits: 
Batal I, Cooper G, Hauskrecht M.
AttachmentSize
PDF icon Batal.pdf962.6 KB
^