FB-STEP: A fuzzy Bayesian network based data-driven framework for spatio-temporal prediction of climatological time series data

作者：

Highlights：

• This work proposes a data-driven framework for spatio-temporal prediction (FB-STEP).

• FB-STEP attempts to address three major challenges in climatological prediction.

• FB-STEP is based on combined fuzzy Bayesian and multifractal analysis technique.

• Validation has been made by predicting climatic condition for five cities in India.

• Study shows improved performance of FB-STEP compared to state-of-the-art methods.

摘要

•This work proposes a data-driven framework for spatio-temporal prediction (FB-STEP).•FB-STEP attempts to address three major challenges in climatological prediction.•FB-STEP is based on combined fuzzy Bayesian and multifractal analysis technique.•Validation has been made by predicting climatic condition for five cities in India.•Study shows improved performance of FB-STEP compared to state-of-the-art methods.

论文关键词：Spatio-temporal analysis,Multivariate prediction,Computational intelligence,Fuzzy Bayesian network,Multifractal analysis,Climatic time series

论文评审过程：Received 20 November 2017, Revised 15 August 2018, Accepted 31 August 2018, Available online 13 September 2018, Version of Record 1 October 2018.

论文官网地址：https://doi.org/10.1016/j.eswa.2018.08.057