Browse Prior Art Database

Multimedia Web Cluster Analysis

IP.com Disclosure Number: IPCOM000105910D
Original Publication Date: 1993-Sep-01
Included in the Prior Art Database: 2005-Mar-20
Document File: 2 page(s) / 48K

Publishing Venue

IBM

Related People

Eisen, IR: AUTHOR [+2]

Abstract

For a new user - new to a complex subject matter - grasping/identifying most significant factors or issues of the subject will typically require a thorough, time consuming study of the subject matter.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 60% of the total text.

Multimedia Web Cluster Analysis

      For a new user - new to a complex subject matter -
grasping/identifying most significant factors or issues of the
subject will typically require a thorough, time consuming study of
the subject matter.

      It is possible to mathematically analyze the graphical
representation of a hyperlinked document.  Performing cluster
analysis of this graphical representation can help identify the most
important concepts or issues of the subject.

      It will be a great help to a new user if he is able to zero in
on the most critical issues fast.  Moreover, in a hypertext document,
the user is more likely to experiment and explore many links.  He is
less likely to follow a particular direction of exploration further
complicating study.

      It should be possible to abstract out from a hyperlinked
document, a graph representing various links in the document.  In
this graph, certain nodes will have a much larger connectivity.

      Moreover, the links in a hypertext document can be assigned
another important attribute:  the attribute of the direction of the
link.  The direction could represent, for example, the causation
relationship of the two nodes which are linked by the link in
question (e.g., slavery as a cause of the Civil War).

      By mathematically analyzing a graph using the direction of each
link and the connectivity of each node, it is possible to perform the
cluster analysis of the graph.  The nodes ...