Neural Architectures for Open-Type Relation Argument Extraction

TL;DR

This paper introduces Open-Type Relation Argument Extraction (ORAE), a new task for extracting non-standard entity arguments from text, and proposes neural models that significantly improve extraction accuracy over baselines.

Contribution

The work defines the ORAE task, creates a distantly supervised dataset, and systematically compares neural architectures, identifying the best model for extracting complex entity arguments.

Findings

Gated recurrent units with CRF achieve the best performance.

Neural models outperform traditional question answering baselines.

A new dataset based on WikiData relations is released.

Abstract

In this work, we introduce the task of Open-Type Relation Argument Extraction (ORAE): Given a corpus, a query entity Q and a knowledge base relation (e.g.,"Q authored notable work with title X"), the model has to extract an argument of non-standard entity type (entities that cannot be extracted by a standard named entity tagger, e.g. X: the title of a book or a work of art) from the corpus. A distantly supervised dataset based on WikiData relations is obtained and released to address the task. We develop and compare a wide range of neural models for this task yielding large improvements over a strong baseline obtained with a neural question answering system. The impact of different sentence encoding architectures and answer extraction methods is systematically compared. An encoder based on gated recurrent units combined with a conditional random fields tagger gives the best results.