Date of Award

2010

Document Type

Campus Access Thesis

Department

Computer Science and Engineering

First Advisor

Michael N. Huhns

Abstract

The goal of this work is to support deductive reasoning over unstructured data called a folksonomy and mapping the information induced by the folksonomy to an ontology. Folksonomy can be described as an implicit structure that emerges from an unstructured collection of keywords.

The concepts that describe and analyze the framework, called Folksonomy Space, support knowledge extraction from a folksonomy are also introduced. The dimensions of Folksonomy Space are explored for amazon.com that will help to delineate salient features of tag sets, tags, taggers, and referenced objects. An analytic framework is presented and the terms and concepts that support exposing the latent information in the folksonomies.

This work proposes several analytic methods and also the development of software techniques to make the implicit structures explicit in a folksonomy. The information induced from the folksonomy can be used for a variety of purposes.

Share

COinS