MGFS: A multi-label graph-based feature selection algorithm via PageRank centrality
作者：
Highlights：
• We have proposed a fast algorithm for feature selection on the multi-label data.
• Features that discriminate classes are linked to provide an undirected weighted graph.
• Features relationships are defined based on correlation distance with labels.
• PageRank algorithm ranks the features according to their importance in weighted graph.
• The proposed multi-label graph based method outperforms competitive methods.
摘要
•We have proposed a fast algorithm for feature selection on the multi-label data.•Features that discriminate classes are linked to provide an undirected weighted graph.•Features relationships are defined based on correlation distance with labels.•PageRank algorithm ranks the features according to their importance in weighted graph.•The proposed multi-label graph based method outperforms competitive methods.
论文关键词：Multi-label feature selection,Correlation distance matrix,Feature-label graph,PageRank centrality
论文评审过程：Received 9 July 2019, Revised 13 October 2019, Accepted 13 October 2019, Available online 16 October 2019, Version of Record 25 October 2019.
论文官网地址：https://doi.org/10.1016/j.eswa.2019.113024
