Feature importance in machine learning models: A fuzzy information fusion approach

Rengasamy, Divish; Mase, Jimiama M.; Kumar, Aayush; Rothwell, Benjamin; Torres, Mercedes Torres; Alexander, Morgan R.; Winkler, David A.; Figueredo, Grazziela P.

Abstract

With the widespread use of machine learning to support decision-making, it is increasingly important to verify and understand the reasons why a particular output is produced. Although post-training feature importance approaches assist this interpretation, there is an overall lack of consensus regarding how feature importance should be quantified, making explanations of model predictions unreliable. In addition, many of these explanations depend on the specific machine learning approach employed and on the subset of data used when calculating feature importance. A possible solution to improve the reliability of explanations is to combine results from multiple feature importance quantifiers from different machine learning approaches coupled with re-sampling. Current state-of-the-art ensemble feature importance fusion uses crisp techniques to fuse results from different approaches. There is, however, significant loss of information as these approaches are not context-aware and reduce several quantifiers to a single crisp output. More importantly, their representation of “importance” as coefficients may be difficult to comprehend by end-users and decision makers. Here we show how the use of fuzzy data fusion methods can overcome some of the important limitations of crisp fusion methods by making the importance of features easily understandable.

Citation

Rengasamy, D., Mase, J. M., Kumar, A., Rothwell, B., Torres, M. T., Alexander, M. R., …Figueredo, G. P. (2022). Feature importance in machine learning models: A fuzzy information fusion approach. Neurocomputing, 511, 163-174. https://doi.org/10.1016/j.neucom.2022.09.053

• Journal Article Type: Article
• Acceptance Date: Sep 4, 2022
• Online Publication Date: Sep 13, 2022
• Publication Date: Oct 28, 2022
• Deposit Date: Sep 16, 2022
• Publicly Available Date: Sep 16, 2022
• Journal: Neurocomputing
• Print ISSN: 0925-2312
• Electronic ISSN: 1872-8286
• Publisher: Elsevier
• Peer Reviewed: Peer Reviewed
• Volume: 511
• Pages: 163-174
• DOI: https://doi.org/10.1016/j.neucom.2022.09.053
• Keywords: Artificial Intelligence; Cognitive Neuroscience; Computer Science Applications
• Public URL: https://nottingham-repository.worktribe.com/output/11199698
• Publisher URL: https://www.sciencedirect.com/science/article/pii/S0925231222011584?via%3Dihub