期刊
TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES
卷 140, 期 -, 页码 -出版社
PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.trc.2022.103709
关键词
Ride-hailing demand prediction; Sparse uncertainty; Hexagonal convolutional long short-term; memory (H-ConvLSTM); Bagging learning
资金
- National Natural Science Foundation of China [51378091, 71871043]
The problem of learning from imbalanced ride-hailing demand data is a new challenge. To achieve better prediction performance, a bagging learning approach based on H-ConvLSTM is proposed, which sets multiple thresholds and selects the best submodel to predict the ride-hailing demand.
The problem of learning from imbalanced ride-hailing demand data with spatiotemporal het-erogeneity and highly skewed demand distributions is a relatively new challenge. Current pre-diction methods usually filter out some spatiotemporal partitions with sparse demands by setting a minimum ride-hailing demand threshold, where the dataset is always assumed to be well balanced in terms of its spatiotemporal partitions, with equal misprediction costs. However, this widely used assumption results in large prediction biases. To achieve better prediction perfor-mance, we propose a bagging learning approach based on hexagonal convolutional long short -term memory (H-ConvLSTM), which combines three components. 1) By setting multiple mini-mum ride-hailing demand thresholds, several subdatasets with different majority ride-hailing demand prediction ranges are obtained. The H-ConvLSTM regression model is applied to each undersampled dataset to train multiple submodels with their respective biased ride-hailing de-mand prediction ranges. 2) The H-ConvLSTM classification model is trained on the total ride-hailing demand dataset to predict the potential demand range for a certain partition at a future time. 3) The submodel with the best performance with respect to the potential demand range is selected to predict the future demand for this partition. Experiments conducted on order data obtained from Didi Chuxing in Chengdu, China, are conducted. The results show that the proposed approach achieves significantly improved prediction performance relative to that of other models.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据