Historical pattern recognition with trajectory similarity for daily tourist arrivals forecasting
作者：
Highlights：
• A forecasting approach based on historical patterns for daily tourist arrivals.
• Two novel trajectory similarity methods for finding historical patterns.
• Similar historical period data is used as input to predict tourist arrivals.
• Empirical results verify that the proposed approach outperforms other benchmarks.
• The forecasting method is excellent and effective on high volatile time series.
摘要
•A forecasting approach based on historical patterns for daily tourist arrivals.•Two novel trajectory similarity methods for finding historical patterns.•Similar historical period data is used as input to predict tourist arrivals.•Empirical results verify that the proposed approach outperforms other benchmarks.•The forecasting method is excellent and effective on high volatile time series.
论文关键词：Daily tourist arrival forecasting,Tourist arrival characteristics analysis,Emergency forecasting,Time series similarity,Trajectory similarity algorithm
论文评审过程：Received 25 November 2021, Revised 26 April 2022, Accepted 26 April 2022, Available online 7 May 2022, Version of Record 10 May 2022.
论文官网地址：https://doi.org/10.1016/j.eswa.2022.117427
