FabricQA-Extractor: A Question Answering System to Extract Information from Documents using Natural Language Questions

TL;DR

FabricQA-Extractor is a question answering system that leverages relational structure knowledge to improve large-scale information extraction from unstructured documents.

Contribution

The paper introduces Relation Coherence, a novel model that exploits relational structure, integrated into FabricQA-Extractor for enhanced extraction performance.

Findings

Relation Coherence improves extraction accuracy.

FabricQA-Extractor performs well on large-scale datasets.

System efficiently handles millions of documents.

Abstract

Reading comprehension models answer questions posed in natural language when provided with a short passage of text. They present an opportunity to address a long-standing challenge in data management: the extraction of structured data from unstructured text. Consequently, several approaches are using these models to perform information extraction. However, these modern approaches leave an opportunity behind because they do not exploit the relational structure of the target extraction table. In this paper, we introduce a new model, Relation Coherence, that exploits knowledge of the relational structure to improve the extraction quality. We incorporate the Relation Coherence model as part of FabricQA-Extractor, an end-to-end system we built from scratch to conduct large scale extraction tasks over millions of documents. We demonstrate on two datasets with millions of passages that Relation Coherence boosts extraction performance and evaluate FabricQA-Extractor on large scale datasets.