Weighted Relaxed Support Vector Machines
Keywords
Classification; Cost-sensitive learning; Imbalanced data; Outliers; Relaxed support vector machines
Abstract
Classification of imbalanced data is challenging when outliers exist. In this paper, we propose a supervised learning method to simultaneously classify imbalanced data and reduce the influence of outliers. The proposed method is a cost-sensitive extension of the relaxed support vector machines (RSVM), where the restricted penalty free-slack is split independently between the two classes in proportion to the number samples in each class with different weights, hence given the name weighted relaxed support vector machines (WRSVM). We compare classification results of WRSVM with SVM, WSVM and RSVM on public benchmark datasets with imbalanced classes and outlier noise, and show that WRSVM produces more accurate and robust classification results.
Publication Date
2-1-2017
Publication Title
Annals of Operations Research
Volume
249
Issue
1-2
Number of Pages
235-271
Document Type
Article
Personal Identifier
scopus
DOI Link
https://doi.org/10.1007/s10479-014-1711-6
Copyright Status
Unknown
Socpus ID
85011407741 (Scopus)
Source API URL
https://api.elsevier.com/content/abstract/scopus_id/85011407741
STARS Citation
Şeref, Onur; Razzaghi, Talayeh; and Xanthopoulos, Petros, "Weighted Relaxed Support Vector Machines" (2017). Scopus Export 2015-2019. 5549.
https://stars.library.ucf.edu/scopus2015/5549