A non-parametric approach to extending generic binary classifiers for multi-classification
作者：
Highlights：
• Ensemble methods combine binary classifiers to yield a multi-classification output.
• One-vs-one ensemble: binary classifiers trained to discriminate each class pair.
• We propose a robust non-parametric probabilistic one-vs-one ensemble method: KDEMRP.
• KDEMRP improves classification performance over state-of-the-art (DCS, DRCW).
• KDEMRP improvements are statistically significant.
摘要
Highlights•Ensemble methods combine binary classifiers to yield a multi-classification output.•One-vs-one ensemble: binary classifiers trained to discriminate each class pair.•We propose a robust non-parametric probabilistic one-vs-one ensemble method: KDEMRP.•KDEMRP improves classification performance over state-of-the-art (DCS, DRCW).•KDEMRP improvements are statistically significant.
论文关键词：Multi-classification,Ensemble method,One-vs-one,Orthogonal subspace,Non-parametric density estimation
论文评审过程：Received 27 May 2015, Revised 3 April 2016, Accepted 13 April 2016, Available online 22 April 2016, Version of Record 26 May 2016.
论文官网地址：https://doi.org/10.1016/j.patcog.2016.04.008
