With the increasing usage of storing textual data into relational databases, there is a demand for the databases to support keyword queries over textual data. Due to the normalization and the inherent connections among tuples in different tables, traditional IR-style ranking and query evaluation methods do not apply. A number of systems have been proposed to deal with this issue. In this thesis, I will give a detailed demonstration and description to our SPARK project. In the project, we study both the effectiveness and the efficiency issues of answering top-k keyword query on a relational database system. We propose a new ranking formula by adapting existing IR techniques on a natural notion of ???virtual document???. Compared with previous approaches, our new ranking method is simple yet effective, and agrees with human being???s perception better. We also study efficient query processing methods based on the new ranking method, and propose algorithms that have minimal accesses to the database. We have conducted extensive experiments on large-scale real databases using two popular RDBMSs. The experimental results demonstrate significant improvement to the alternative approaches in terms of both retrieval effectiveness and efficiency. We build a prototype of SPARK system on top of popular RDBMS based on these new techniques to satisfy different kinds of users and to support various query modes.
| Identifer | oai:union.ndltd.org:ADTP/258483 |
| Date | January 2009 |
| Creators | Luo, Yi , Computer Science & Engineering, Faculty of Engineering, UNSW |
| Publisher | Awarded by:University of New South Wales. Computer Science & Engineering |
| Source Sets | Australiasian Digital Theses Program |
| Language | English |
| Detected Language | English |
| Rights | Copyright Luo Yi ., http://unsworks.unsw.edu.au/copyright |
Page generated in 0.0023 seconds