Data mining for state space orthogonalization in adaptive dynamic programming

作者：

Highlights：

• Data mining techniques are employed to orthogonalize a multicollinear state space.

• Orthogonalization facilitates design and analysis of computer experiments.

• Partial least squares combined with stepwise regression is recommended.

• A case study addresses multicollinearity between ozone pollution concentrations.

摘要

•Data mining techniques are employed to orthogonalize a multicollinear state space.•Orthogonalization facilitates design and analysis of computer experiments.•Partial least squares combined with stepwise regression is recommended.•A case study addresses multicollinearity between ozone pollution concentrations.

论文关键词：Data mining,Design and analysis of computer experiments,Approximate dynamic programming,Ozone pollution

论文评审过程：Received 15 September 2016, Revised 2 January 2017, Accepted 23 January 2017, Available online 28 January 2017, Version of Record 7 February 2017.

论文官网地址：https://doi.org/10.1016/j.eswa.2017.01.020