An online support vector machine for the open-ended environment

作者：

Highlights：

• An online learning architecture for Support Vector Machine is proposed.

• Learn representative prototypes from the input data in an online incremental way.

• Works effectively when new stream data and novel class data come.

• Much faster than the state-of-the-art SVM and can handle much larger scale problems.

摘要

•An online learning architecture for Support Vector Machine is proposed.•Learn representative prototypes from the input data in an online incremental way.•Works effectively when new stream data and novel class data come.•Much faster than the state-of-the-art SVM and can handle much larger scale problems.

论文关键词：Online learning,Open-ended learning environment,Support vector machine,Novel classes,Stream data

论文评审过程：Received 24 January 2018, Revised 28 September 2018, Accepted 14 October 2018, Available online 13 November 2018, Version of Record 19 November 2018.

论文官网地址：https://doi.org/10.1016/j.eswa.2018.10.027