PUP

This is the official implementation of our ICDE'20 paper:

Yu Zheng, Chen Gao, Xiangnan He, Yong Li, Depeng Jin, Price-aware Recommendation with Graph Convolutional Networks, In Proceedings of IEEE ICDE 2020.

First download the Yelp dataset (link) and the category file (link).

Then generate training data and test data using the codes in src/yelp_restaurant_pipeline:

python generate_action_log.py
python generate_bridge.py
python generate_training_data_sample_fm_bpr.py

Then start the visdom server:

visdom -port 33332

Then simply run the following command to reproduce the experiments:

python app.py --flagfile ./config/yelp.cfg

If you use our codes and datasets in your research, please cite:

@inproceedings{zheng2020price,
  title={Price-aware recommendation with graph convolutional networks},
  author={Zheng, Yu and Gao, Chen and He, Xiangnan and Li, Yong and Jin, Depeng},
  booktitle={2020 IEEE 36th International Conference on Data Engineering (ICDE)},
  pages={133--144},
  year={2020},
  organization={IEEE}
}