QA4IE: A Question Answering based Framework for Information Extraction

TL;DR

QA4IE introduces a question answering-based framework for extracting high-quality relation triples across sentences, overcoming limitations of traditional IE methods in handling cross-sentence and informal relations.

Contribution

The paper presents a novel QA-based IE framework and a large benchmark dataset with extensive relation types and high-quality annotations.

Findings

QA4IE outperforms baseline IE systems on the benchmark

The framework effectively extracts cross-sentence relation triples

The benchmark contains 293K documents and 2M relation triples

Abstract

Information Extraction (IE) refers to automatically extracting structured relation tuples from unstructured texts. Common IE solutions, including Relation Extraction (RE) and open IE systems, can hardly handle cross-sentence tuples, and are severely restricted by limited relation types as well as informal relation specifications (e.g., free-text based relation tuples). In order to overcome these weaknesses, we propose a novel IE framework named QA4IE, which leverages the flexible question answering (QA) approaches to produce high quality relation triples across sentences. Based on the framework, we develop a large IE benchmark with high quality human evaluation. This benchmark contains 293K documents, 2M golden relation triples, and 636 relation types. We compare our system with some IE baselines on our benchmark and the results show that our system achieves great improvements.