Topic network: a semantic model for effective learning

There has been tremendous interest in sharing and retrieving information through the Web. A search engine can be used to retrieve relevant web documents. However, the sheer volume of results returned often requires substantial effort to determine which documents are relevant, what information they contain, and how they relate to each other. Learning about a particular topic could be facilitated if it were possible to automatically find and summarize the important topics in a given domain. This can be achieved by defining a learning model that is based on the automated analysis of the importance of topics and relationships between them. In this thesis, an intelligent and dynamic model called Topic Network is proposed. Given a topic of interest, Topic Network generates a network of relevant terms and associations between them based on available Web resources. The steps involved are (1) Retrieving data from the Web and Ontologies, (2) Selecting relevant terms and their relationships using association and importance factors, and (3) Visualizing the network and associated web documents for each topic in the network. A visual prototype system useful for clinical trial research has been developed for three important domains such as Clinical Trials, Informed Consent, and Generalized Anxiety Disorder. We have evaluated the model in terms of accuracy and performance as well as performed a comparison with a traditional learning model. The results illustrate the enhanced efficiency and quality of learning through the Web when using the Topic Network model.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Masters

Discipline thesis:degree_discipline

Computer Science (UMKC)

Grantor dc:publisher

University of Missouri--Kansas City

Year dc:date.issued

2011

Author dc:creator

• Garg, Taru, 1987-

Advisor dc:contributor.advisor

• Lee, Yugyung, 1960-

Language dc:language.iso

en_US

Handle dc:identifier.uri

http://hdl.handle.net/10355/10898

OAI identifier oai:identifier

oai:mospace.umsystem.edu:10355/10898