DeepSTD: Mining Spatio-temporal Disturbances of Multiple Context Factors for Citywide Traffic Flow Prediction

<p align="center"> <img width="500" height="400" src=./figure/DeepSTD.png> </p>

This is the implementation of DeepSTD in the following paper:
Chuanpan Zheng, Xiaoliang Fan, Chengwu Wen, Longbiao Chen, Cheng Wang, and Jonathan Li. "DeepSTD: Mining Spatio-temporal Disturbances of Multiple Context Factors for Citywide Traffic Flow Prediction", published in IEEE Transactions on Intelligent Transportation Systems (T-ITS).

Data

The Chengdu dataset with 500m*500m grid size and 15-minute time interval is provided in the './data' folder.

Requirements

Python 3.7.10, tensorflow 1.14.0, numpy 1.16.4, pandas 0.24.2

Results

We provide a pre-trained model, which achieve the following performance:

| Chengdu | Workday | Weekend | All days | | -------------- | ---------- | ------------ | -------- | | DeepSTD | 10.36 | 10.78 | 10.48 |

Citation

If you find this repository useful in your research, please cite the following paper:

@article{ DeepSTD:TITS,
  author   = "Chuanpan Zheng and Xiaoliang Fan and Chenglu Wen and Longbiao Chen and Cheng Wang and Jonathan Li"
  title    = "DeepSTD: Mining Spatio-temporal Disturbances of Multiple Context Factors for Citywide Traffic Flow Prediction",
  journal  = "IEEE Transactions on Intelligent Transportation Systems",
  volume   = "21",
  number   = "9",
  pages    = "3744--3755",
  year     = "2020"
}