Previous MSc Theses

"View Maintenance of RDF Structures." S. M. S. Shah. J. Wilson. Department of Computer and Information Sciences, University of Strathclyde. 2013. Download PDF (BibTeX) MSc Advanced Computer Science

Abstract:
The Semantic Web is an extension of current Web technologies that aims to transform the way the Web functions. The ultimate goal of the Semantic Web is data integration and sharing of information on the Internet. Most of the present Web content is meant for human consumption, computers are unable to understand the meaning of the content. Semantic Web provides a complete framework for creating a new generation of Web content that is also intelligible by machines. At the heart of the Semantic Web is the RDF (Resource Description Framework) which has been recommended by W3C (World Wide Web Consortium) as the new universal data format for information interchange on the internet. RDF is a structured data format and solves the problems of existing data representation formats. The current practice of updating RDF data involves regenerating the underlying structure after modifications, which is expensive in terms of time and computational resources. As more and more Web content is created using RDF, there is a need for simple and efficient mechanisms to manage voluminous RDF structures.

This project aims to assess the performance of update operations on RDF structures through views. Views are a powerful way to manipulate a section of data from the underlying database. A view is subset of the underlying database, when changes are made to the underlying database; the views also need to be updated. RDF statements are made up of three components i.e. subject, predicate and object. In relational terms, RDF structures are three dimensional databases. The goal of this project is to assess performance of view maintenance of RDF structures i.e. to update RDF structures through views and measure performance of insert and delete operations with respect to the size of the RDF structures. A prototype application has been developed to measure the performance through experiments. The results show that as size increases, the time of insert operations increases proportionally where as for delete operations time remains stable.