PhD Thesis Proposal Defence


"Towards Fully Automatic Data Integration for Web Databases"

by

Mr. Weifeng Su


Abstract:

An important part of today's Web is Web databases, in which 80% of the
databases are structured databases. To facilitate a user to retrieve
relevant records from different Web databases simultaneously, we propose
a meta-querying system comprised of three components: query interface
integrator, data extractor and result integrator. In each component, a
novel method is proposed that tries to address the respective problem
automatically. In the query interface integrator, the attribute
occurrence patterns in multiple query interfaces are used to find the
attributes that match in different interfaces. In the data extractor, a
domain ontology is first learned from the information overlap in the
query results from different web databases and it is then used to extract
the data encoded in the HTML pages. In the result integrator, to identify
the duplicates that exist in the results from different web databases, a
set of negative records is first constructed by assuming that the records
from the same web database are unique and then, starting from the
negative records, an iterative algorithm identifies the duplicates from
different web databases. Preliminary experimental results show that these
automatic, novel methods can achieve very high precision and outperform
existing methods.


Date:     		Wednesday, 24 January 2007

Time:                   2:00p.m.-4:00p.m.

Venue:                  Room 3501
			lifts 25-26

Committee Members:      Prof. Frederick Lochovsky (Supervisor)
			Prof. Dik-Lun Lee (Chairperson)
			Dr. Qiong Luo
			Dr. Wilfred Ng


**** ALL are Welcome ****