Network intrusion detection using equality constrained-optimization-based extreme learning machines

作者：

Highlights：

• Applying the equality constrained-optimization-based extreme learning machine to network intrusion detection.

• An adaptively incremental learning strategy is proposed to derive the optimal number of hidden neurons.

• The optimization criteria and a way of adaptively increasing hidden neurons are developed.

• The proposed approach is effective in building models with good attack detection rates and fast learning speed.

摘要

•Applying the equality constrained-optimization-based extreme learning machine to network intrusion detection.•An adaptively incremental learning strategy is proposed to derive the optimal number of hidden neurons.•The optimization criteria and a way of adaptively increasing hidden neurons are developed.•The proposed approach is effective in building models with good attack detection rates and fast learning speed.

论文关键词：Supervised learning,Extreme learning machine,Adaptively incremental learning,Support vector machine,Network intrusion detection

论文评审过程：Received 5 July 2017, Revised 1 February 2018, Accepted 4 February 2018, Available online 8 February 2018, Version of Record 28 February 2018.

论文官网地址：https://doi.org/10.1016/j.knosys.2018.02.015