A robust SVM-based approach with feature selection and outliers detection for classification problems
作者：
Highlights：
• A new model to classify data based on Support vector machines is introduced.
• The model deals with outlier detection and feature selection simultaneously.
• Strategies for initializing the big M parameters of the model are developed.
• An heuristic algorithm based on kernel search to compute the classifier is presented.
• The classification performance of the model is tested on real-life datasets.
摘要
•A new model to classify data based on Support vector machines is introduced.•The model deals with outlier detection and feature selection simultaneously.•Strategies for initializing the big M parameters of the model are developed.•An heuristic algorithm based on kernel search to compute the classifier is presented.•The classification performance of the model is tested on real-life datasets.
论文关键词：Data science,Classification,Support vector machine,Outliers detection,Feature selection,Mixed integer programming
论文评审过程：Received 29 May 2020, Revised 10 October 2020, Accepted 6 April 2021, Available online 20 April 2021, Version of Record 5 May 2021.
论文官网地址：https://doi.org/10.1016/j.eswa.2021.115017
