FS-MLC: Feature selection for multi-label classification using clustering in feature space
作者：
Highlights：
• A novel method, named FS-MLC, for feature selection in Multi-label classification is proposed.
• FS-MLC uses clustering to find the similarity among features.
• It is a wrapper method that does not require generating the number of feature subsets linearly proportional to the number of labels in the dataset.
• It reduces the dimensionality of the dataset up to 23%-93%.
• The experimental results show an impressive prediction performance improvement of the associated classifiers.
摘要
•A novel method, named FS-MLC, for feature selection in Multi-label classification is proposed.•FS-MLC uses clustering to find the similarity among features.•It is a wrapper method that does not require generating the number of feature subsets linearly proportional to the number of labels in the dataset.•It reduces the dimensionality of the dataset up to 23%-93%.•The experimental results show an impressive prediction performance improvement of the associated classifiers.
论文关键词：Multi-label learning,Feature selection,Wrapper method,Instance space,Feature space
论文评审过程：Received 19 October 2019, Revised 27 February 2020, Accepted 9 March 2020, Available online 23 March 2020, Version of Record 23 March 2020.
论文官网地址：https://doi.org/10.1016/j.ipm.2020.102240
