/OAG-entity-alignment

Primary LanguagePythonMIT LicenseMIT

OAG Entity Alignment Task

Prerequisites

  • Linux
  • Python 3
  • NVIDIA GPU + CUDA cuDNN

Getting Started

Installation

Clone this repo.

git clone https://github.com/zfjsail/OAG-entity-alignment.git
cd OAG-entity-alignment

Please install dependencies by

pip install -r requirements.txt

Dataset

The dataset can be downloaded from BaiduPan (with password 445n) or Aliyun. Unzip the file and put the data directory into project directory.

How to run

cd $project_path
export CUDA_VISIBLE_DEVICES='?'  # specify which GPU(s) to be used
python processing.py  # all pre-processing process
python train_rnn_match.py --n-try 5 --entity-type author  # RNN-based matching model for author alignment

If you want to know about all supported baselines, see run.sh for details. For entity types, we support three types of entities (affliation, venue, and author) for entity alignment. For matching models, we support RNN-based, CNN-based, and heterogeneous graph attention network (HGAT) based models.