Knowledge-Driven Cross-Document Relation Extraction

TL;DR

This paper introduces KXDocRE, a novel method for cross-document relation extraction that integrates domain knowledge, enhances interpretability, and improves performance in identifying relationships across multiple documents.

Contribution

The paper presents a new framework, KXDocRE, that incorporates domain knowledge into cross-document relation extraction, providing interpretability and better accuracy than existing methods.

Findings

KXDocRE outperforms baseline models in relation extraction accuracy.

The method provides explanatory text for predicted relations.

Incorporating domain knowledge improves the interpretability of the model.

Abstract

Relation extraction (RE) is a well-known NLP application often treated as a sentence- or document-level task. However, a handful of recent efforts explore it across documents or in the cross-document setting (CrossDocRE). This is distinct from the single document case because different documents often focus on disparate themes, while text within a document tends to have a single goal. Linking findings from disparate documents to identify new relationships is at the core of the popular literature-based knowledge discovery paradigm in biomedicine and other domains. Current CrossDocRE efforts do not consider domain knowledge, which are often assumed to be known to the reader when documents are authored. Here, we propose a novel approach, KXDocRE, that embed domain knowledge of entities with input text for cross-document RE. Our proposed framework has three main benefits over baselines: 1) it incorporates domain knowledge of entities along with documents' text; 2) it offers interpretability by producing explanatory text for predicted relations between entities 3) it improves performance over the prior methods.