This is the official implementation of An Attribute-Driven Mirror Graph Network for Session-based Recommendation from SIGIR 2022.
- Python=3.7
- PyTorch=1.10.1
- numpy=1.20.3
Our data has been preprocessed and is available at https://www.dropbox.com/sh/h548vcds8a4m3qs/AABH-YavkkoNFPR_RMTtmILOa?dl=0. You need to download the datasets folder and put it under the root. All the original dataset files are available at https://www.dropbox.com/sh/dbzmtq4zhzbj5o9/AAAMMlmNKL-wAAYK8QWyL9MEa/Datasets?dl=0&subfolder_nav_tracking=1.
- Diginetica dataset has a category attribute. Its attribute information can be found at https://competitions.codalab.org/competitions/11161#learn_the_details-data2.
- 30music dataset has an artist attribute. Each item's attribute value can be found in the original dataset file.
- The brief version of the Tmall dataset is also available on that website. And you can find its corresponding original file containing attribute information at https://tianchi.aliyun.com/dataset/dataDetail?dataId=42. It includes a category and a brand attribute.
Train and evaluate the model:
python main.py --dataset TmallThe sequence length of Diginetica dataset is relatively long. You can also train the model by utilizing 2 GPUs:
python main_dual_GPU.py --dataset diginetica
@inproceedings{DBLP:conf/sigir/LaiMZLWS22,
author = {Siqi Lai and
Erli Meng and
Fan Zhang and
Chenliang Li and
Bin Wang and
Aixin Sun},
editor = {Enrique Amig{\'{o}} and
Pablo Castells and
Julio Gonzalo and
Ben Carterette and
J. Shane Culpepper and
Gabriella Kazai},
title = {An Attribute-Driven Mirror Graph Network for Session-based Recommendation},
booktitle = {{SIGIR} '22: The 45th International {ACM} {SIGIR} Conference on Research
and Development in Information Retrieval, Madrid, Spain, July 11 -
15, 2022},
pages = {1674--1683},
publisher = {{ACM}},
year = {2022},
url = {https://doi.org/10.1145/3477495.3531935},
doi = {10.1145/3477495.3531935},
timestamp = {Fri, 08 Jul 2022 17:25:07 +0200},
biburl = {https://dblp.org/rec/conf/sigir/LaiMZLWS22.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}