Markov blanket-based universal feature selection for classification and regression of mixed-type data
作者：
Highlights：
• Development of a universal feature selection method for mixed-type data is urgent.
• A generalized conditional independence test is proposed for any type of data.
• New Markov blanket feature selection with the test is proposed for mixed-type data.
• The method can be applied to both regression and classification of mixed-type data.
• The method outperforms other comparable methods in terms of accuracy and sparseness.
摘要
•Development of a universal feature selection method for mixed-type data is urgent.•A generalized conditional independence test is proposed for any type of data.•New Markov blanket feature selection with the test is proposed for mixed-type data.•The method can be applied to both regression and classification of mixed-type data.•The method outperforms other comparable methods in terms of accuracy and sparseness.
论文关键词：Markov blanket,Multivariate feature selection,Conditional independence test,Likelihood-ratio test,Classification,Regression
论文评审过程：Received 25 July 2018, Revised 5 March 2020, Accepted 20 March 2020, Available online 14 May 2020, Version of Record 23 June 2020.
论文官网地址：https://doi.org/10.1016/j.eswa.2020.113398
