An effective multivariate time series classification approach using echo state network and adaptive differential evolution algorithm

作者：

Highlights：

• Propose a new multivariate time series classification approach (CADE).

• Incorporate an efficient ADE for optimal decision of parameters in learning phase.

• CADE outperforms other approaches in terms of accuracy for most cases.

• CADE shows good robustness for all the cases.

摘要

•Propose a new multivariate time series classification approach (CADE).•Incorporate an efficient ADE for optimal decision of parameters in learning phase.•CADE outperforms other approaches in terms of accuracy for most cases.•CADE shows good robustness for all the cases.

论文关键词：Multivariate time series classification,Recurrent neural network,Adaptive differential evolution algorithm

论文评审过程：Received 15 June 2015, Revised 24 August 2015, Accepted 28 August 2015, Available online 8 September 2015, Version of Record 20 October 2015.

论文官网地址：https://doi.org/10.1016/j.eswa.2015.08.055