Semantic Web

Ontology-Based Information Retrieval Model for the Semantic Web

Song Jun-feng, Zhang Wei-ming, Xiao Wei-dong, Li Guo-hui, Xu Zhen-ning

(School of Information System and Management, The National University of Defense Technology,

Changsha 410073,China)

sjfemail@hotmail.com

Abstract

In this paper, we describe ontology-based information

retrieval model for the Semantic Web. By using OWL Lite

as standard ontology language, which is a suitable

tradeoff between expressivity of knowledge and

complexity of reasoning problems, ontology is generated

through translating and integrating domain ontologies.

The terms defined in ontology are used as metadata to

markup the Web’s content; these semantic markups are

semantic index terms for information retrieval. We can

obtain the equivalent classes of semantic index terms by

using description logic reasoner. The logical views of

documents and user information needs, generated in

terms of the equivalent classes of semantic index terms,

can represent documents and user information needs well,

so the performance of information retrieval can be

improved effectively when suitable ranking function is

chosen. The practicability of this model is discussed.

Finally, the related works are introduced, and the

conclusion and our future work are given.

1. Introduction

The research on information retrieval (IR) is boosting

with the coming of Information Age. Everyday, a lot of

data are produced. The need for retrieving useful

information from the huge data sets is urgent.

In this paper, retrieval unit is document, which can be

text, image, video and audio. Information Retrieval [1] is

to find all the relevant documents, which satisfies the user

information need, from a document collection.

According to the research on information retrieval

models, we know that at present, vector model is

dominant thought among researchers, practitioners and

the Web community. [1] But information retrieval for the

Web is...