Keyword Query Expansion on Linked Data Using Linguistic and Semantic Features

Effective search in structured information based on textual user input is of high importance in thousands of applications. Query expansion methods augment the original query of a user with alternative query elements with similar meaning to increase the chance of retrieving appropriate resources. In this work, we introduce a number of new query expansion features based on semantic and linguistic inferencing over Linked Open Data. We evaluate the effectiveness of each feature individually as well as their combinations employing several machine learning approaches. The evaluation is carried out on a training dataset extracted from the QALD question answering benchmark. Furthermore, we propose an optimized linear combination of linguistic and lightweight semantic features in order to predict the usefulness of each expansion candidate. Our experimental study shows a considerable improvement in precision and recall over baseline approaches.

I. Introduction

With the growth of the Linked Open Data cloud, a vast amount of structured information was made publicly available. Querying that huge amount of data in an intuitive way is challenging. SPARQL, the standard query language of the semantic web, requires exact knowledge of the vocabulary and is not accessible by laypersons. Several tools and algorithms have been developed that make use of semantic technologies and background knowledge [10], such as TBSL [17], SINA [16] and WolframlAlpha¹

http://www.wolframalpha.com

. Those tools suffer from a mismatch between query formulation and the underlying knowledge base structure that is known as the vocabulary problem [8]. For instance, using DBpedia as knowledge base, the query “Who is married to Barack Obama?” could fail, because the desired property in DBpedia is labeled “spouse” and there is no property labeled “married to”.