Q-Traffic is a large-scale traffic prediction dataset consisting of three subsets:

• Query sub-dataset: collected from Baidu Maps, including about 114 million user queries. Each query records the start timestamp, start location coordinates, destination coordinates, and estimated travel time (minutes).
• Traffic speed sub-dataset: Traffic speed data in the same area and time period as the query sub-dataset. Contains 15073 road sections, covering approximately 738.91 kilometers.
• Road network sub-dataset: Provides the start node (snode) and end node (enode) for each road segment in the traffic speed sub-dataset, and builds the road network topology based on it. In addition, various geographical attributes (such as width, length, speed limit and number of lanes) and social attributes (such as weekdays, weekends, public holidays, peak hours and off-peak hours) of each road segment are provided.

Citation

@inproceedings{bbliaojqZhangKDD18deep, title = {Deep Sequence Learning with Auxiliary Information for Traffic Prediction}, author = {Binbing Liao and Jingqing Zhang and Chao Wu and Douglas McIlwraith and Tong Chen and Shengwen Yang and Yike Guo and Fei Wu}, booktitle = {Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining}, pages = {537–546}, year = {2018}, organization = {ACM} }