Theoretical and empirical analysis of filter ranking methods: Experimental study on benchmark DNA microarray data

作者：

Highlights：

• A study on applicability and effect of filter ranking methods on microarray data.

• 10 microarray datasets (both binary class and multi-class) with varying dimension.

• Concluded that out of all the methods Mutual Information (MI) gives the best results.

• An informed choice about selecting an appropriate filtering method for their work.

摘要

•A study on applicability and effect of filter ranking methods on microarray data.•10 microarray datasets (both binary class and multi-class) with varying dimension.•Concluded that out of all the methods Mutual Information (MI) gives the best results.•An informed choice about selecting an appropriate filtering method for their work.

论文关键词：Feature selection,Filter method,Microarray data,Cancer classification,Gene expression

论文评审过程：Received 22 May 2020, Revised 28 October 2020, Accepted 8 December 2020, Available online 16 December 2020, Version of Record 8 January 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.114485