The combination of context information to enhance simple question answering

TL;DR

This paper explores how incorporating context information like entity type and outdegree can improve the accuracy of simple factoid question answering systems based on knowledge bases.

Contribution

It introduces a question answering system that leverages context information to enhance fact selection for single-relation questions.

Findings

Context information improves question answering accuracy

Entity notable type and outdegree are effective features

Experimental results confirm the benefits of context integration

Abstract

With the rapid development of knowledge base,question answering based on knowledge base has been a hot research issue. In this paper, we focus on answering singlerelation factoid questions based on knowledge base. We build a question answering system and study the effect of context information on fact selection, such as entity's notable type,outdegree. Experimental results show that context information can improve the result of simple question answering.