Enhanced fuzzy time series forecasting model based on hesitant differential fuzzy sets and error learning

作者：

Highlights：

• An enhanced fuzzy time series model is developed.

• Fuzzy silhouette criterion is adopted to determine the number and length of fuzzy intervals.

• Aggregation operator is utilized to aggregate hesitant information of fuzzy set.

• Differential fuzzy sets are included in the model to reconnoiter deep trends.

• An optimal error learning mechanism is considered to enhance the forecasting performance.

摘要

•An enhanced fuzzy time series model is developed.•Fuzzy silhouette criterion is adopted to determine the number and length of fuzzy intervals.•Aggregation operator is utilized to aggregate hesitant information of fuzzy set.•Differential fuzzy sets are included in the model to reconnoiter deep trends.•An optimal error learning mechanism is considered to enhance the forecasting performance.

论文关键词：Fuzzy time series,Fuzzy silhouette,Hesitant information,Differential fuzzy set,Error learning

论文评审过程：Received 13 May 2020, Revised 6 August 2020, Accepted 24 September 2020, Available online 6 October 2020, Version of Record 26 October 2020.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.114056