Improved support vector machine algorithm for heterogeneous data

作者：

Highlights：

• We propose an algorithm to map nominal features to a numerical space via minimizing estimated generalization errors.

• We integrate the mapping algorithm with support vector machines and result in an improved learning algorithm from heterogeneous data.

• Experiments show the proposed technique is effective for learning with heterogeneous data and also help deal with imbalanced tasks.

摘要

Highlights•We propose an algorithm to map nominal features to a numerical space via minimizing estimated generalization errors.•We integrate the mapping algorithm with support vector machines and result in an improved learning algorithm from heterogeneous data.•Experiments show the proposed technique is effective for learning with heterogeneous data and also help deal with imbalanced tasks.

论文关键词：Support vector machine,Heterogeneous data,Nominal attribute,Numerical attribute,Classification learning

论文评审过程：Received 17 October 2013, Revised 2 November 2014, Accepted 17 December 2014, Available online 24 December 2014.

论文官网地址：https://doi.org/10.1016/j.patcog.2014.12.015