/User-Difference-Attention

Primary LanguagePython

User-Difference-Attention

Implementation of our paper "UDA: A User-Difference Attention for Group Recommendation". Our implementation is based on Attentive-Group-Recommendation.

Please cite our paper if you use our codes. Thanks!

BibTeX:

@article{ZAN2021401,
title = {UDA: A user-difference attention for group recommendation},
journal = {Information Sciences},
volume = {571},
pages = {401-417},
year = {2021},
issn = {0020-0255},
doi = {https://doi.org/10.1016/j.ins.2021.04.084},
author = {Shuxun Zan and Yujie Zhang and Xiangwu Meng and Pengtao Lv and Yulu Du},
}

Getting Started

1. Environment Settings

  • python 3.6
  • basic python packages
pip install tqdm==4.59.0 scipy==1.5.4 numpy==1.16.4
  • pytorch 1.6.0+cu101 (GPU version. Take cu101 for example.)
pip install torch==1.6.0+cu101 -f https://download.pytorch.org/whl/torch_stable.html

2. Examples to run the models

(1) ML100K dataset

uda:

python uda_main.py --dataset ml100k_1000_3 --embedding_size 128

agree:

python agree_main.py --dataset ml100k_1000_3 --embedding_size 128

(2) Last.fm dataset

uda:

python uda_main_lastfm.py --dataset lastfm_2 --embedding_size 128 --num_negatives 5  --lr "[1e-3,5e-4,1e-4]"

agree:

python agree_main_lastfm.py --dataset lastfm_2 --embedding_size 128 --num_negatives 5  --lr "[1e-3,5e-4,1e-4]"

(3) CAMRa2011 dataset can be found in Attentive-Group-Recommendation.

Related Projects

I also contribute to these open-source projects: