PhD Qualifying Examination


"A Survey on Distributed Web Information Retrieval"


By Miss Jing Zhao


Abstract:

The tremendous size and rapid growing pace of World Wide Web makes
search engines crucial for people to seek desired information. A
search engine claws and indexes the Web, parses user queries, and
retrieves relevant Web resources following its ranking criterion.
The dynamic and distributed nature of Web resources, however,
makes it infeasible for a single search engine to index the whole
Web while keep its database up-to-date. Centralized architecture
cannot adapt to the inherent characteristics of World Wide Web.

Distributed search systems offer attractive advantages over
centralized ones, where the search tasks are performed in a
decentralized network, and each node indexes an assigned region of
the Web. The distributed architecture brings with information
retrieval a promise of harnessing idle storage, computation and
network resources, and excellent scalability due to elimination of
bottleneck at certain nodes. While distributed search is gaining
more and more attentions, it also faces challenges such as network
instability, insecurity, and unreliability of individual sources,
concerning which many efforts have been taken by previous
researchers.

This survey reviews Web information retrieval (IR) technology,
with a great focus on the area of distributed search systems. Some
of the design and implementation issues of distributed search
systems are discussed. We also made a taxonomy of distributed
information retrieval architectures and did a case study, where
some of the existing or new distributed search systems are
investigated.


Date:     		Friday, 19 December 2003

Time:                   10:00a.m.-12:00noon

Venue:                  Room 1403
			lifts 17-18

Committee Members:      Prof. Dik-Lun Lee (Supervisor)
			Dr. Qiong Luo (Supervisor)
                        Prof. Frederick Lochovsky (Chairperson)
			Prof. Hongjun Lu
			Dr. Dimitris Papadias


**** ALL are Welcome ****