基于Tucker分解和季节性自回归移动平均模型的出租车出行需求预测

打开文本图片集

中图分类号：TP391 文献标志码：A

Taxi Travel Demand Forecasting Based on Tucker Decomposition and Seasonal Autoregressive Integrated Moving Average Model

CHU Ben-jia，YAN Hong-yu，LI Jian-bo （Collge of Computer Science& Technology，Qingdao University，Qingdao 266071，China）

Abstract： To enhance the accuracy and efficiency of taxi trip demand forecasting，a model combining Tucker decomposition and seasonal autoregressive moving average model was proposed. The spatiotemporal modeling of the core tensor after Tucker decomposition was carried out to better capture the internal multi-modal structure and spatiotemporal correlation of taxi travel demand， so as to improve the prediction ability of the model. The tensor representation of taxi demand data was constructed，the Tucker decomposition was used to extract core features，and the seasonal autoregressive integrated moving average model was used for forecasting. Experimental results show that the proposed method has better accuracy and computational efficiency compared with the baseline model.

Keywords： travel demand forecasting； tucker decomposition； seasonal autoregressive integrated moving average model

出租车是城市交通的重要组成之一[1]，乘客可以随时随地乘坐出租车，不需要提前规划或等待。（剩余9460字）