Improved Neural Relation Detection for Knowledge Base Question Answering

TL;DR

This paper introduces a hierarchical residual neural network for improved relation detection in KBQA, significantly enhancing accuracy and achieving state-of-the-art results on multiple benchmarks.

Contribution

It presents a novel deep residual bidirectional LSTM model for relation detection and integrates it into a KBQA system, advancing the state of the art.

Findings

Achieves superior relation detection performance.

Enables the KBQA system to reach state-of-the-art accuracy.

Effective in both single-relation and multi-relation QA benchmarks.

Abstract

Relation detection is a core component for many NLP applications including Knowledge Base Question Answering (KBQA). In this paper, we propose a hierarchical recurrent neural network enhanced by residual learning that detects KB relations given an input question. Our method uses deep residual bidirectional LSTMs to compare questions and relation names via different hierarchies of abstraction. Additionally, we propose a simple KBQA system that integrates entity linking and our proposed relation detector to enable one enhance another. Experimental results evidence that our approach achieves not only outstanding relation detection performance, but more importantly, it helps our KBQA system to achieve state-of-the-art accuracy for both single-relation (SimpleQuestions) and multi-relation (WebQSP) QA benchmarks.