An Automatic Graph Construction Framework based on Large Language Models for Recommendation

Introduction

This is the pytorch implementation of AutoGraph proposed in the paper An Automatic Graph Construction Framework based on Large Language Models for Recommendation.

Required Packages

pip install torch --index-url https://download.pytorch.org/whl/cu118

pip install dgl-cu118 -f https://data.dgl.ai/wheels/repo.html

pip install vllm transformers vector-quantize-pytorch

Data preprocess

Download the original dataset, put them under directory 'Datasets', and run the notebooks in Preprocess.

LLM Inference

Inference on user preference and item knowledge.

python gen_knowledge.py

Encode LLM Knowledge

Encode the inferenced knowledges into vectors.

python gen_embedding.py

Quantization

python quantization.py

Enhance REC Model

python main.py --plus_gnn_embed