GraphIE: A Graph-Based Framework for Information Extraction

TL;DR

GraphIE is a novel graph-based framework for information extraction that leverages non-local dependencies to improve prediction accuracy across various domains.

Contribution

It introduces a graph convolutional approach to incorporate broad contextual dependencies in IE tasks, surpassing traditional sequence tagging models.

Findings

Outperforms state-of-the-art sequence tagging models

Effective across textual, social media, and visual IE tasks

Provides richer representations through graph convolutions

Abstract

Most modern Information Extraction (IE) systems are implemented as sequential taggers and only model local dependencies. Non-local and non-sequential context is, however, a valuable source of information to improve predictions. In this paper, we introduce GraphIE, a framework that operates over a graph representing a broad set of dependencies between textual units (i.e. words or sentences). The algorithm propagates information between connected nodes through graph convolutions, generating a richer representation that can be exploited to improve word-level predictions. Evaluation on three different tasks --- namely textual, social media and visual information extraction --- shows that GraphIE consistently outperforms the state-of-the-art sequence tagging model by a significant margin.