Learning Relational Dependency Networks for Relation Extraction

TL;DR

This paper introduces a relation extraction method using Relational Dependency Networks that integrates various features and supervision techniques, achieving competitive results in the KBP 2015 benchmark.

Contribution

It presents a novel pipeline employing RDNs for relation extraction, incorporating weak supervision, word embeddings, joint learning, and human advice.

Findings

RDNs effectively model diverse features

Achieves competitive performance on KBP 2015

Demonstrates flexibility in incorporating supervision and features

Abstract

We consider the task of KBP slot filling -- extracting relation information from newswire documents for knowledge base construction. We present our pipeline, which employs Relational Dependency Networks (RDNs) to learn linguistic patterns for relation extraction. Additionally, we demonstrate how several components such as weak supervision, word2vec features, joint learning and the use of human advice, can be incorporated in this relational framework. We evaluate the different components in the benchmark KBP 2015 task and show that RDNs effectively model a diverse set of features and perform competitively with current state-of-the-art relation extraction.