CNN-based Relation Extraction

About

CNN-based Relation Extraction Model

prerequisite

• java 1.8
• maven

How to use

refer edu.kaist.mrlab.nn.pcnn.Main.java and PCNN.conf files

How to run

mvn exec:java -Dexec.args="--training"
or
mvn exec:java -Dexec.args="--testing"

Data Example

라타쿵가	에콰도르	country	코토팍시 국제공항(, )은  << _obj_ >>   << 코토팍시_주 >>   << _sbj_ >> 에 있는  << 국제공항 >> 이다.
코토팍시_주	에콰도르	country	코토팍시 국제공항(, )은  << _obj_ >>   << _sbj_ >>   << 라타쿵가 >> 에 있는  << 국제공항 >> 이다. 

TSV format; subject, object, relation, sentence with two position of target entities;
<< >>: entity boundary
_sbj_, _obj_: special token for subject and object entities

Licenses

• CC BY-NC-SA Attribution-NonCommercial-ShareAlike
• If you want to commercialize this resource, please contact to us

Maintainer

Sangha Nam nam.sangha@kaist.ac.kr

Publisher

Machine Reading Lab @ KAIST

Citation

@article{nam2018distant, title={Distant Supervision for Relation Extraction with Multi-sense Word Embedding}, author={Nam, Sangha and Han, Kijong and Kim, Eun-kyung and Choi, Key-Sun}, journal={Global Wordnet Conference, Workshops on Wordnets and Word Embeddings}, year={2018} }

Acknowledgement

This work was supported by Institute for Information & communications Technology Promotion(IITP) grant funded by the Korea government(MSIT) (2013-0-00109, WiseKB: Big data based self-evolving knowledge base and reasoning platform)