Hybrid feature selection using component co-occurrence based feature relevance measurement

作者：

Highlights：

• A component co-occurrence based feature relevance measurement is proposed.

• An effective and efficient hybrid feature selection frame is proposed.

• A feature weight based union operation is proposed to merge the feature subsets.

• The traditional HAC algorithm is improved to increase the running speed.

摘要

•A component co-occurrence based feature relevance measurement is proposed.•An effective and efficient hybrid feature selection frame is proposed.•A feature weight based union operation is proposed to merge the feature subsets.•The traditional HAC algorithm is improved to increase the running speed.

论文关键词：Feature selection,Mutual information,Hierarchical agglomerative clustering,Support vector machine,K-nearest neighbor

论文评审过程：Received 25 October 2017, Revised 5 January 2018, Accepted 26 January 2018, Available online 31 January 2018, Version of Record 19 March 2018.

论文官网地址：https://doi.org/10.1016/j.eswa.2018.01.041