An Improved IAMB Algorithm for Markov Blanket Discovery

Abstract

Finding an efficient way to discover Markov blanket is one of the core issue s in data mining. This paper first discusses the problems existed in IAMB algorithm which is a typical algorithm for discovering the Markov b lanket of a target variable from the training dat a, and then proposes an improved algorithm λ -IAMB based on the improving approach which contains two aspects: code optimization and the improving strategy for conditional independence testing. E xperimental results show that λ -IAMB algorithm performs better than IAMB by finding Markov blanket of variables in typical Bayesian network and by testing the performance of them as feature selection method on some well-known real world datasets .