Title: " Personalized Web: P2P Overlay Architecture for Personalized Resource Discovery, Access, and Sharing over the Internet "

When: Thursday November 18, 2004 , 2:00pm - 4:00pm

Where: Room 503, IS Building

Who: Chatree Sangpachatanaruk

Committee:

Dr. Taieb Znati, DIST, Committee Chair
Dr. Martin Weiss, DIST
Dr. Michael Spring, DIST
Dr. Sujata Banerjee, HP Lab
Dr. Kirk Pruhs, Department of Computer Science

Abstract:Due to its exponential growth and decentralized nature, the Internet has evolved into a chaotic repository, making it difficult for users to discover and access resources of interest. The emergence of Semantic Web, however, provides Internet users with the ability to associate detailed semantics with resources. This will in turn facilitate the development of ontology-based resource discovery tools to help users retrieve information in an efficient manner. It is widely believed, however, that the Semantic Web of the future will be a complex web of small ontologies, largely created by groups of web users, and ruled by the same sort of anarchy that currently rules the rest of the web.

A central argument of this proposal is that information retrieval and knowledge sharing in an Internet environment will be most effective if the users' personal interests are taken into account. In support of the above argument, the concept of a Personalized Web is proposed to enable a user-defined, structured access to Internet resources. It provides the basis for a flexible and effective platform for the development of an autonomous personalization system which will allow users to create their own personalized views of the Internet.  

The main goal of this proposal is to develop the data structures, mechanisms and protocols to support the functionalities of a Personalized Web, in a scalable and robust manner. The approach taken to achieve this goal, leverage the benefits of two technologies, P2P overlay networking agent-enabled Semantic Web. In this proposal, we investigate the design goals of a Personalized Web architecture, and the investigate the performance requirements of its basic functionalities and services. We then discuss the main challenges to be addressed for the development of scalable and robust Personalized Web. We finally articulate a research agenda to enable a Personalized Web and assess its performance.