Gene selection for microarray data classification via multi-objective graph theoretic-based method
作者：
Highlights：
• A novel gene selection method has been developed by integrating the concept of node centrality and community detection.
• The main goal of this method is to select a subset of the genes with lowest similarity and highest dependency.
• In this proposed method the optimal number of final gene set is determined automatically.
• Experimental results showed that the proposed method has the best performance among different gene selection methods.
• The results on five microarray datasets indicate that this method improves microarray data classification accuracy.
摘要
•A novel gene selection method has been developed by integrating the concept of node centrality and community detection.•The main goal of this method is to select a subset of the genes with lowest similarity and highest dependency.•In this proposed method the optimal number of final gene set is determined automatically.•Experimental results showed that the proposed method has the best performance among different gene selection methods.•The results on five microarray datasets indicate that this method improves microarray data classification accuracy.
论文关键词：Microarray data classification,Gene selection,Feature selection,Community detection,Node centrality,Multi-objective
论文评审过程：Received 31 October 2020, Revised 23 November 2021, Accepted 27 November 2021, Available online 3 December 2021, Version of Record 6 December 2021.
论文官网地址：https://doi.org/10.1016/j.artmed.2021.102228
