Detecting mobike hotspots in new cities by city domain adaptation model.
Tensorflow implementation for the paper 'Where Will Dockless Shared Bikes be Stacked? — Parking Hotspots Detection in a New City', which is publised on the applied data science track of KDD 2018 (The 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining).
We aim at transferring knowlege from the city popular with dockess shared bike to the new cities and detecting potential hotspots roads before expanding to these cities.
We collect muliple geo-related data from differet sources, including:
- Mobike data (by crawler);
- POI data from (Baidu Map);
- Satellite light (Light Polution);
- Road network (OpenStreetMap);
- Transportation hubs (collect manually);
- Business centers (Fliggy);
If you're interested in these data, you can refer to the data/meta_data
folder, we provide the part of the preprocessed subway station, business center, satellite light cluster results data from Shanghai, Beijing, and Ningbo three cities. You can also crawl more data from our provided sources.
The project is organized as follows:
data/
meta_data: multiple source data;
metrics/
the maximum mean discrepancy metric implementation
models/
core models and run scripts
results/
the spatial-temporal characteristics analysis results
road_match/
map matching and feature extraction scripts
util/
general common functions
If you use this code or the data for your research, please cite our paper, paper link:
@inproceedings{Liu2018Where,
title={Where Will Dockless Shared Bikes be Stacked? — Parking Hotspots Detection in a New City},
author={Liu, Zhaoyang and Shen, Yanyan and Zhu, Yanmin},
booktitle={The 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining},
pages={378-386},
year={2018},
}
-
Our previous work on dockless shared bike distribution inferences, paper link:
@inproceedings{Liu2018Inferring, title={Inferring Dockless Shared Bike Distribution in New Cities}, author={Liu, Zhaoyang and Shen, Yanyan and Zhu, Yanmin}, booktitle={Eleventh ACM International Conference on Web Search and Data Mining}, pages={378-386}, year={2018}, }
-
Bike lane planning work in KDD 2017, paper link:
@inproceedings{Bao2017Planning, title={Planning Bike Lanes based on Sharing-Bikes' Trajectories}, author={Bao, Jie and He, Tianfu and Ruan, Sijie and Li, Yanhua and Zheng, Yu}, booktitle={ACM SIGKDD International Conference on Knowledge Discovery and Data Mining}, pages={1377-1386}, year={2017}, }
-
Unsupervised Domain Adaptation by Backpropagation, paper link:
@article{Ganin2014Unsupervised, title={Unsupervised Domain Adaptation by Backpropagation}, author={Ganin, Yaroslav and Lempitsky, Victor}, pages={1180-1189}, year={2014}, }
- For academic and non-commercial use only.
- For commercial use, please contact Mobike Company