期刊
ANALYTICA CHIMICA ACTA
卷 1192, 期 -, 页码 -出版社
ELSEVIER
DOI: 10.1016/j.aca.2021.339352
关键词
LIBS; Classification; Feature importance; SVM; Interpretable machine learning
资金
- Czech Grant Agency [20-19526Y]
- Faculty of Mechanical Engineering of Brno University of Technology [FSI-S-20-6353]
- Research Infrastructure RECETOX RI - Ministry of Education, Youth and Sport [LM2018121]
- Operational Programme Research, Development and Innovation e project CETOCOEN EXCELLENCE [CZ.02.1.01/0.0/17_043/0009632]
This article presents four approaches to interpret support vector machines (SVMs) and investigates the classification task of 19 algal and cyanobacterial species. The study finds that different feature importance metrics provide complementary information, and identifies the SVM model's bias towards features with a large variance.
Laser-induced breakdown spectroscopy is often combined with a multivariate black box model-such as support vector machines (SVMs)-to obtain desirable quantitative or qualitative results. This approach carries obvious risks when practiced in high-stakes applications. Moreover, the lack of understanding of a black-box model limits the user's ability to fine-tune the model. Thus, here we present four approaches to interpret SVMs through investigating which features the models consider important in the classification task of 19 algal and cyanobacterial species. The four feature importance metrics are compared with popular approaches to feature selection for optimal SVM performance. We report that the distinct feature importance metrics yield complementary and often comparable information. In addition, we identify our SVM model's bias towards features with a large variance, even though these features exhibit a significant overlap between classes. We also show that the linear and radial basis kernel SVMs weight the same features to the same degree. (c) 2021 Elsevier B.V. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据