Skip to content

Latest commit

 

History

History
39 lines (24 loc) · 1.19 KB

README.md

File metadata and controls

39 lines (24 loc) · 1.19 KB
Raw

CasMLN

LLM-enhanced Cascaded Multi-level Learning on Temporal Heterogeneous Graphs (CasMLN)

This is the code associated with the paper "LLM-enhanced Cascaded Multi-level Learning on Temporal Heterogeneous Graphs" accepted by SIGIR 2024. Our code references DHGAS. Thanks for their great contributions!

Dependencies

Install dependencies (with python >= 3.8):

pip install torch-scatter torch-sparse torch-geometric -f https://data.pyg.org/whl/torch-1.12.1+cu116.html
pip install networkx
pip install openai

Install this repo as a library:

pip install -e .

Datasets

All the processed datasets we used in the paper can be downloaded at Baidu Yun (password:lmrb). Put datasets in the folder 'cmln/data' to run experimments.

Run scripts

To run the code, you must set your own openai.api_key and openai.api_base in the file cmln/model/LLM.py.

python scripts/run/run_model.py --dataset Aminer
python scripts/run/run_model.py --dataset Ecomm
python scripts/run/run_model.py --dataset Yelp-nc
python scripts/run/run_model.py --dataset covid