Investigating ontology based query expansion using a probabilistic retrieval model

Bhogal, Jagdev (2011). Investigating ontology based query expansion using a probabilistic retrieval model. (Unpublished Doctoral thesis, City University London)

[img]
Preview
PDF
Download (3MB) | Preview

Abstract

This research briefly outlines the problems of traditional information retrieval systems and discusses the different approaches to inferring context in document retrieval. By context we mean word disambiguation which is achieved by exploring the generalisation-specialisation hierarchies within a given ontology. Specifically, we examine the use of ontology based query expansion for defining query context. Query expansion can be done in many ways and in this work we consider the use of relevance feedback and pseudo-relevance feedback for query expansion. We examine relevance feedback and pseudo-relevance to ascertain the existence of performance differences between relevance feedback and pseudo-relevance feedback. The information retrieval system used is based on the probabilistic retrieval model and the query expansion method is extended using information from a news domain ontology. The aim of this project is to assess the impact of the use of the ontology on the query expansion results. Our results show that ontology based query expansion has resulted in a higher number of relevant documents being retrieved compared to the standard relevance feedback process. Overall, ontology based query expansion improves recall but does not produce any significant improvements for the precision results. Pseudo-relevance feedback has achieved better results than relevance feedback. We also found that reducing or increasing the relevance feedback parameters (number of terms or number of documents) does not correlate with the results. When comparing the effect of varying the number of terms parameter with the number of documents parameter, the former benefits the pseudo-relevance feedback results but the latter has an additional effect on the relevance feedback results. There are many factors which influence the success of ontology based query expansion. The thesis discusses these factors and gives some guidelines on using ontologies for the purpose of query expansion.

Item Type: Thesis (Doctoral)
Subjects: Z Bibliography. Library Science. Information Resources > Z665 Library Science. Information Science
Divisions: School of Informatics > Department of Information Science
City University London PhD theses
URI: http://openaccess.city.ac.uk/id/eprint/2946

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year

View more statistics