Year

2008

Degree Name

Master of Computer Science

Department

School of Computer Science and Software Engineering

Abstract

The explosion of World Wide Web has brought endless amount of information within our reach. In order to take advantage of this phenomenon, text search becomes a major contemporary research challenge. Due to the nature of the Web, assisting users to find desired information is still a challenging task. In this thesis, we investigate semantic anlaysis techniques which can facilitate the search process at semantic level. We also study the problem that short queries are less informative and difficult to convey the user's intention into the search service system. We propose a generalized framework to address these issues. We conduct a case study of movie plot search in which a semantic analyzer seamlessly works with a user's intention detector. Our experimental results show the importance and effectiveness of intention detection and semantic analysis techniques.

Download

02Whole.pdf (10724 kB)

Share

COinS
 

Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.