Compressed Interaction Graph based Framework for Muti-behavior Recommendation

This repository contains TensorFlow codes and datasets for the paper.

Environment

The codes of CIGF are implemented and tested under the following development environment:

• python=3.6.12
• tensorflow=1.14.0
• numpy=1.16.0
• scipy=1.5.2

Datasets

We utilized three datasets to evaluate CIGF: Beibei, Tmall, and IJCAI Contest. The purchase behavior is taken as the target behavior for all datasets. The last target behavior for the test users are left out to compose the testing set. We filtered out users and items with too few interactions.

Just Run It！

• Beibei

python base_framework_origin_final.py --data beibei 

• Tmall

python base_framework_origin_final.py --data tmall --gnn_layer 4

• IJCAI

python base_framework_samp_final.py --data ijcai

Citation

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

@inproceedings{cigf,
  title={Compressed Interaction Graph based Framework for Multi-behavior Recommendation},
  author={Guo, Wei and Meng, Chang and Yuan, Enming and He, Zhicheng and Guo, Huifeng and Zhang, Yingxue and Chen, Bo and Hu, Yaochen and Tang, Ruiming and Li, Xiu and others},
  booktitle={Proceedings of the ACM Web Conference 2023},
  pages={960--970},
  year={2023}
}