Multi-view Inference for Relation Extraction with Uncertain Knowledge

TL;DR

This paper presents a multi-view inference framework that leverages uncertain knowledge graphs to enhance relation extraction, demonstrating improved performance on sentence- and document-level tasks.

Contribution

It introduces a novel multi-view inference approach that integrates local context and global uncertain knowledge from KGs into relation extraction models.

Findings

Achieves competitive results on relation extraction benchmarks.

Validates the effectiveness of uncertain knowledge in RE.

Demonstrates the benefit of multi-view inference framework.

Abstract

Knowledge graphs (KGs) are widely used to facilitate relation extraction (RE) tasks. While most previous RE methods focus on leveraging deterministic KGs, uncertain KGs, which assign a confidence score for each relation instance, can provide prior probability distributions of relational facts as valuable external knowledge for RE models. This paper proposes to exploit uncertain knowledge to improve relation extraction. Specifically, we introduce ProBase, an uncertain KG that indicates to what extent a target entity belongs to a concept, into our RE architecture. We then design a novel multi-view inference framework to systematically integrate local context and global knowledge across three views: mention-, entity- and concept-view. The experimental results show that our model achieves competitive performances on both sentence- and document-level relation extraction, which verifies the effectiveness of introducing uncertain knowledge and the multi-view inference framework that we design.