Regularized Attentive Capsule Network for Overlapped Relation Extraction

TL;DR

This paper introduces RA-CapNet, a novel capsule network with attention and regularization, to improve the extraction of overlapped relations in noisy distantly supervised datasets, significantly enhancing accuracy.

Contribution

The paper proposes a regularized attentive capsule network with multi-head attention and disagreement regularization for better overlapped relation extraction.

Findings

Achieves significant improvements on widely used datasets.

Effectively identifies multiple relations in noisy data.

Outperforms existing relation extraction models.

Abstract

Distantly supervised relation extraction has been widely applied in knowledge base construction due to its less requirement of human efforts. However, the automatically established training datasets in distant supervision contain low-quality instances with noisy words and overlapped relations, introducing great challenges to the accurate extraction of relations. To address this problem, we propose a novel Regularized Attentive Capsule Network (RA-CapNet) to better identify highly overlapped relations in each informal sentence. To discover multiple relation features in an instance, we embed multi-head attention into the capsule network as the low-level capsules, where the subtraction of two entities acts as a new form of relation query to select salient features regardless of their positions. To further discriminate overlapped relation features, we devise disagreement regularization to explicitly encourage the diversity among both multiple attention heads and low-level capsules. Extensive experiments conducted on widely used datasets show that our model achieves significant improvements in relation extraction.