jackfrost168 / CF_Diff

CF-Diff

This is the pytorch implementation of our paper at SIGIR 2024:

Collaborative Filtering Based on Diffusion Models: Unveiling the Potential of High-Order Connectivity

Hou, Yu and Park, Jin-Duk and Shin, Won-Yong

The implementation of diffusion model and evaluation parts are referred to DiffRec. Thank you for this contribution.

Environment

• Anaconda 3
• python 3.8.17
• pytorch 1.13.1
• numpy 1.24.3
• math

Usage

Data

The user-item interactions, train/valid/test, are in './datasets' folder. "sec_hop_inters_ML_1M.pt" contains the information of second-hop user-item interactions and "multi_hop_inters_ML_1M.pt" contains multi-hop user-item interactions. More data about "high-order interactions" can be found here. More "saved_models" can be found here.

Training

CF-Diff

cd ./CF_Diff
python main.py

Inference

cd ./CF_Diff
python inference.py

Citation

@inproceedings{hou2024collaborative,
title = {Collaborative Filtering Based on Diffusion Models: Unveiling the Potential of High-Order Connectivity},
author = {Hou, Yu and Park, Jin-Duk and Shin, Won-Yong},
booktitle = {Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval},
year = {2024}
}