Robust support vector data description for outlier detection with noise or uncertain data
作者:
Highlights:
• We propose two new SVDD models which improve the robustness to noise.
• Cutoff distance-based local density can mitigate the effect of noise towards SVDD.
• Tolerated gap of SVDD with ε-insensitive loss can improve generalization performance.
摘要
•We propose two new SVDD models which improve the robustness to noise.•Cutoff distance-based local density can mitigate the effect of noise towards SVDD.•Tolerated gap of SVDD with ε-insensitive loss can improve generalization performance.
论文关键词:Outlier detection,Support vector data description,Local density,ε-insensitive loss
论文评审过程:Received 11 July 2015, Revised 23 September 2015, Accepted 25 September 2015, Available online 9 October 2015, Version of Record 8 November 2015.
论文官网地址:https://doi.org/10.1016/j.knosys.2015.09.025