Web Mining
Web Data Mining, drawing from its conventional counterpart,
refers to the extraction of previously unknown, implicit
and potentially useful strategic information from
the WWW. In Whoweda, we explore basic operations that
constitute web data mining and defines a framework
to mine a set of semi-structured, related websites.
We also investigate some of the important issues governing
knowledge discovery on the web that distinguish it
from ordinary data mining.
Web Bags
A web bag is a "web table" which allows
multiple occurrences of "identical" web
tuples. A web bag helps to discover knowledge related
to query traversed path, visible documents or web
sites, luminous documents or web sites, etc. Some
of these knowledge can be used further in refining
a user's query. There are a number of new challenges
related to web bag due to the richer nature of the
Whoweda data model which handles unstructured data.
What exactly is a web bag in Whoweda? How are web
bags created in Whoweda? Is there is a need to materialize
web bags? What is the usefulness of web bags from
the perspective of information provided to users?
In this work, we address some of these challenges
Resercher
Dr.
Sanjay Madria
Dr. Sourav Bhowmic, Nanyang Technological University,
Singapore
Dr. Ng Wee Keong, Nanyang Technological University, Singapore
|