Feature weighting methods: A review
作者：
Highlights：
• An extensive review of Feature Weighting methods based on a new global taxonomy.
• A new global taxonomy focused on the learning strategy, methodology and feedback.
• A classification based on the input data, application and ML performance measure.
• Considerations and guide points for selecting the FW strategy.
• A collection of unsolved challenges for future work in this scientific field.
摘要
•An extensive review of Feature Weighting methods based on a new global taxonomy.•A new global taxonomy focused on the learning strategy, methodology and feedback.•A classification based on the input data, application and ML performance measure.•Considerations and guide points for selecting the FW strategy.•A collection of unsolved challenges for future work in this scientific field.
论文关键词：Feature weighting,Feature importance,Feature relevance,Review
论文评审过程：Received 11 May 2020, Revised 15 September 2020, Accepted 10 June 2021, Available online 17 June 2021, Version of Record 25 June 2021.
论文官网地址：https://doi.org/10.1016/j.eswa.2021.115424
