KDocRE

KDocRE:Knowledge-Driven Cross-Document Relation Extraction

##Accepted in ACL 2024, findings##

Acknowledgments

This repository contains code adapted from the following research papers for the purpose of cross document-level relation extraction. We extend our gratitude to the authors for generously sharing their clean and valuable code implementations.

1. Entity-centered Cross-document Relation Extraction

• Paper: Entity-centered Cross-document Relation Extraction

• Authors: Wang, Fengqi and Li, Fei and Fei, Hao and Li, Jingye and Wu, Shengqiong and Su, Fangfang and Shi, Wenxuan and Ji, Donghong and Cai, Bo",Wang Xu, Kehai Chen, Tiejun Zhao

• Year: 2022

• **Code implementation can be found here: [https://github.com/MakiseKuurisu/ecrim]

Requirements

• Follow the guideliens from : https://github.com/thunlp/CodRED
• pip install torch-geometric
• pip install a2t
• sudo apt install redis-server
• start Redis-server using: sudo service redis-server start

Alternative using Docker

• docker build -t doc:v0 .

• docker run --gpus all -it -d --shm-size=20gb --name=doc doc:v0

• login Docker and install the following:

• apt-get update &&
  apt-get -y install sudo

• pip install torch-geometric

• pip install a2t

• sudo apt install redis-server

• start Redis-server using: sudo service redis-server start

• Navigate to data/rawdata folder

• wget https://thunlp.oss-cn-qingdao.aliyuncs.com/wiki_ent_link.jsonl

• wget https://thunlp.oss-cn-qingdao.aliyuncs.com/distant_documents.jsonl

• wget https://thunlp.oss-cn-qingdao.aliyuncs.com/popular_page_ent_link.jsonl

• Navigate to data directory:

• python3 load_data_doc.py

• python3 redis_doc.py

Directory structure

In this directory structure, you have a folder named "C" containing a subdirectory "code." Within the "code" directory, there are several files and subdirectories:

• data: Directory to store data
• context: Files for creating context.
• r/: Directory to store model checkpoints.
• main.py: File for training the code.
• explanation_withrelevance.py: File to generate explanation.

Datasets

This project utilizes the following datasets:

• CoDRED Dataset: The DocRED dataset can be accessed ( https://github.com/thunlp/CodRED) place in data/ directory

Training

Follow the steps below to start the training process:

1. Train reasoning module: Navigate to the KDocRE directory using the following command:

   cd/ KDocRE bash train.sh

Testing

Navigate to the KDocRE directory using the following command:

cd/ KDocRE bash test.sh

Explanation

• explaination is generated in explanation.txt
• explaination with relevance is stored in relevance.txt

Citation

Please cite:

@misc{jain2024knowledgedriven,
      title={Knowledge-Driven Cross-Document Relation Extraction}, 
      author={Monika Jain and Raghava Mutharaju and Kuldeep Singh and Ramakanth Kavuluru},
      year={2024},
      eprint={2405.13546},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}