Self Contrastive Learning for Session-based Recommendation

This repository provides the code for the paper titled Self Contrastive Learning for Session-based Recommendation, making the integration of our code contributions into other projects more accessible.

Quick Links

• Self Contrastive Learning for Session-based Recommendation
  • Quick Links
  • Overview
  • 1. Requirements and Installation
  • 2. Prepare the datasets
  • 3. Run our code
  • Bugs or questions?
  • Citation
  • Acknowledgement

Overview

You can reproduce the experiments of our paper Self Contrastive Learning for Session-based Recommendation. We implement three baseline approaches, including

• Global Context Enhanced Graph Neural Networks for Session-based Recommendation, SIGIR 2020
• Self-Supervised Graph Co-Training for Session-based Recommendation, CIKM 2021
• Self-Supervised Hypergraph Convolutional Networks for Session-based Recommendation, AAAI 2021

and evaluate them on three datasets, including TMALL, diginetica, and Nowplaying.

1. Requirements and Installation

Please refer to the repository of each baseline approach (GCE-GNN, COTREC, and DHCN) for the installation and requirements.

2. Prepare the datasets

We provide datasets in the data folder in each baseline folder, including GCE-GNN, COTREC, and DHCN.

3. Run our code

Please refer to the README.md in each baseline folder (GCE-GNN, COTREC, and DHCN) for the instructions to run the code.

Bugs or questions?

If you have any questions regarding the code or the paper, please feel free to reach out to Zhengxiang at [email protected]. If you experience any difficulties while using the code or need to report a bug, feel free to open an issue. We kindly ask that you provide detailed information about the problem to help us provide effective support.

Citation

@inproceedings{shi2023self,
title = {Self Contrastive Learning for Session-based Recommendation},
author = {Shi, Zhengxiang and Xi, Wang and Lipani, Aldo},
publisher = {Springer},
address = {Glasgow, Scotland},
booktitle={European Conference on Information Retrieval (ECIR 2024)},
url = {https://arxiv.org/abs/2306.01266},
year = {2023},
}

Acknowledgement

This repository is built upon the following repositories:

• GCE-GNN
• COTREC
• DHCN