A multiple criteria active learning method for support vector regression

作者：

Highlights：

• We present an active learning method in the context of ε-insensitive SVR.

• The proposed method exploits relevancy, diversity and density criteria.

• The three criteria are implemented according to the SVR properties.

• All three criteria are applied in two consecutive steps.

• Experimental results show the robustness of the proposed technique.

摘要

•We present an active learning method in the context of ε-insensitive SVR.•The proposed method exploits relevancy, diversity and density criteria.•The three criteria are implemented according to the SVR properties.•All three criteria are applied in two consecutive steps.•Experimental results show the robustness of the proposed technique.

论文关键词：Regression,Parameters estimation,Active learning,Support vector regression

论文评审过程：Received 5 August 2013, Revised 31 January 2014, Accepted 4 February 2014, Available online 13 February 2014.

论文官网地址：https://doi.org/10.1016/j.patcog.2014.02.001