PLSPL

Code for Personalized long-and short-term preference learning for next POI recommendation. TKDE'20

Environment:

• Python
• Pytorch

Data preprocess

The sequence is preprocessed by a sliding window. The length of each session is 20. The last one is the ground truth. You can change the setting based on your model.

Run data preprocess:

python preprocess_longshort.py

In the preprocess.py, I also add the preprocessing for other compared baselines. You can omit them if you don't use them.

Run

python train_long_short.py

Cite

If you use the code, please cite the following paper:

@article{wu2020personalized,
  title={Personalized long-and short-term preference learning for next POI recommendation},
  author={Wu, Yuxia and Li, Ke and Zhao, Guoshuai and Xueming, QIAN},
  journal={IEEE Transactions on Knowledge and Data Engineering},
  year={2020},
  publisher={IEEE}
}

@inproceedings{wu2019long,
  title={Long-and short-term preference learning for next POI recommendation},
  author={Wu, Yuxia and Li, Ke and Zhao, Guoshuai and Qian, Xueming},
  booktitle={Proceedings of the 28th ACM international conference on information and knowledge management},
  pages={2301--2304},
  year={2019}
}