KDDC

This is the official implementation of our paper titled "KDDC: Knowledge-Driven Disentangled Causal Metric Learning for Pre-Travel Out-of-Town Recommendation", which has been accepted by the IJCAI 2024 conference.

Pytorch versions are provided.

Pytorch: https://pytorch.org

Data

We have released the travel behavior dataset Foursquare and Yelp which are generated based on the Foursquare and Yelp dataset. You can run the model with these out-of-town data provided in the respective folder.

Requirements

• python 3.x
• paddle 2.x / torch >= 1.7
• pgl / dgl>=0.6

Run Our Model

Simply run the following command to train and evaluate:

cd ./code
python main.py --ori_data {...} --dst_data {...} --trans_data {...} --pp_graph_path {...} ---save_path {...} --mode train --kg --train_trans --dc --infer --dm

Citation

If you find this work helpful, please cite our paper:

BibTeX

@inproceedings{liu2024kddc,
  title={KDDC: Knowledge-driven disentangled causal metric learning for pre-travel out-of-town recommendation},
  author={Liu, Yinghui and Shen, Guojiang and Cui, Chengyong and Zhao, Zhenzhen and Han, Xiao and Du, Jiaxin and Zhao, Xiangyu and Kong, Xiangjie},
  booktitle={Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence, Jeju Island, Republic of Korea},
  pages={4--9},
  year={2024}
}