Mapping Relational Databases to Semantic Web Ontological Model-Space Project Management case study

by Wondu Mallede and Supervisors (Prof Farhi Marir, Dr Vassil Vassilev, Dr Yanguo Jing)

Abstract: The current web experience gives us a fairly abundant data. Using a few keywords and common search engines, it usually doesn't fail to return search results as well. With all its openness, the web gives anyone a chance to contribute ideas to be shared by the whole world about any topic. But the web often feels like it is a mile wide, but an inch deep. How can we build a more integrated, consistent, deep web experience (Allemang & Hendler, 2008)? This PhD work address this problem of accessing deep web (particularly structured databases) by developing models and ontology for mapping relational database structure into a semantic web representation.

Semantic Web sets the standards for a better data representation and semantics that helps us use the computer to do all the semantic analysis rather than search using keywords and read through each search result one by one from the screen. This machine-readable data model that is represented by a Uniform Resource Identifier (URI) will make the data not only readable to the naked eye but also understandable by computers.

Mapping existing relational databases, which are locked behind in the deep web and only accessible through web services and application interfaces, to Semantic web schema makes the data universal, interoperable and machine processable. The use of domain specific ontology makes our mapping process geared to satisfy the interpretation and conceptualisation of the domain knowledge which ultimately helps the efficiency of mapping relational database to Semantic knowledge base.

The mapping implementation involves Resource Description Format (RDF), reasoning logic, Rule Interchange Format (RIF) etc. The research project focuses on how to take advantage of the already existing vast relational data to build a knowledge base in a semantic format (RDF). In the current Semantic architecture RDF is a favoured data serialization format to represent and manage knowledge which is imprecise, uncertain, partially true and approximate. The ultimate goal of the research is a formal heuristics-based methodology for mapping relational database to semantic knowledge base.

Keywords: Semantic Web, Semantic Mapping, Semantic Knowledge Management, RDF, RIF

Short Biographies:

Mr. Wondu Mallede (PhD student) obtained M.Sc in Database Systems from University of Westminster in 2007. He is currently registered as a research student at London Metropolitan University and also working as a database application developer using Oracle development tools and Java. His interests include Semantic web, databases and query languages.

Prof. Farhi Marir (MSc, PhD, FBCS, CITP) is Professor of computer science & director of the Knowledge Management Research Centre in the Faculty of Computing (FoC). He led more than 15 EU & UK funded research projects. Three of them related to this project: OSCON project (DoE funded) for the development of a multi-dimension model for construction to allow each construction application (AutoCAD, Surveying tools, etc...) rendering construction object in their own application interface, CBRefurb (EPSRC funded) for the development of an adaptive user interface for evaluating the cost of refurbishing large construction building using case-based reasoning methods and VIGILANT (private funding) for developing an intelligent user interface for content and event based indexing and retrieval car park CCTV video clips. He also led 10 PhDs to completion: three of them developed content-based indexing and retrieval algorithms for text and multimedia documents and the fourth developed a new framework for e-business processes into e-workflow management systems.

Dr. Vassil T. Vassilev, MBCS, Assoc. Prof. and Reader in Computer Science. Senior Lecturer at the Faculty of Computing and member of ISRC. PhD in Artificial Intelligence. Current interests in Semantic Web.

Dr. Yanguo Jing is a Principal Lecturer in Computing at the London Metropolitan University. Yanguo has a BSc (Hons, 1st class) and MSc in Computer Science from Dalian Maritime University, China and a PhD in Computer Science from Heriot-Watt University. His PhD thesis is in the area of User Modelling and Intelligent User Interfaces. Current research interests are in the areas of Web and Mobile Technologies.