Active cleaning of label noise
作者：
Highlights：
• Novel method for label noise removal from data is introduced.
• It significantly reduces the required number of examples to be reviewed.
• Support vectors of SVM classifier can capture around 99% of label noise examples.
• Two-class SVM captures more label noise examples than one-class SVM classifier
• Combination of one-class and two-class SVM produces a marginal improvement.
摘要
Highlights•Novel method for label noise removal from data is introduced.•It significantly reduces the required number of examples to be reviewed.•Support vectors of SVM classifier can capture around 99% of label noise examples.•Two-class SVM captures more label noise examples than one-class SVM classifier•Combination of one-class and two-class SVM produces a marginal improvement.
论文关键词：Support vectors,Label noise,Mislabeled examples
论文评审过程：Received 20 May 2015, Revised 16 August 2015, Accepted 17 September 2015, Available online 30 September 2015, Version of Record 27 November 2015.
论文官网地址：https://doi.org/10.1016/j.patcog.2015.09.020
