Status: Past project

Project Description:
Ontologies have been widely accepted as the most advanced knowledge representation model. They are among the most important building blocks of semantic web, hence, very crucial for the success of semantic web. Huge effort is needed from the domain expert in order to construct ontologies manually. There is a need for semi-automatic approach in ontology building which will help the domain expert in constructing extensive domain ontologies efficiently. We propose the use of text mining techniques, especially mining the domain specific texts and glossaries/dictionaries in order to find groups of concepts/terms which are related to each other. Such groups of related concepts/terms will enable the domain expert to either, evaluate and update the existing ontology in case those concepts are already defined in the ontology, or to enrich the existing ontology in case those concepts are not defined. This will be an iterative refinement process with the newly available knowledge bases and/or domain specific texts or glossaries.

Start Date: October 2003

End Date: December 2004

Affiliated Faculty:
Tim Finin
Jin-Ping (Jack) Gwo

Students:
Viral Parekh

Tags: ontology, text mining, semantic web, learning

There is 1 associated publication:  Hide the list...

1 Refereed Publication

2004

1. Viral Parekh et al., "Mining Domain Specific Texts and Glossaries to Evaluate and Enrich Domain Ontologies", InProceedings, International Conference of Information and Knowledge Engineering, June 2004, 2824 downloads.

There are 0 associated resources:  Hide the list...

Research Areas:
 Knowledge Representation and Reasoning
 Semantic Web