Date of Graduation

12-2011

Document Type

Thesis

Degree Name

Bachelor of Science

Degree Level

Undergraduate

Department

Computer Science and Computer Engineering

Advisor

Gauch, Susan

Reader

Deaton, Russell J.

Second Reader

Parkerson, James Patrick

Abstract

Website personalization systems seek to give users unique, tailored content and experiences on the Internet. A key feature of these systems is a user profile that represents each user in a way that distinguishes them from others. In current personalization systems, the data used to create these profiles is extremely limited, which leads to a host of problems and ineffectual personalization. The main goal of this thesis is to improve these personalization systems by addressing their lack of data and its poor quality, breadth, and depth. This is accomplished by analyzing and classifying the content of each user's Internet browsing activity, rather than just their activity on a single website, to autonomously build persistent, ontology-based user profiles. Furthermore, these profiles are built and stored on a remote server, which allows them to be easily made available to approved websites in the interest of providing the data to enable accurate, relevant, and up-to-date personalization.

Share

COinS