Re: DM: Datamining in Internet...

Pablo,

Web mining is comprised of three somewhat interdependent
application areas: usage mining, structure mining and
content mining. I think you are inquiring about structure
mining, which refers to the application of data
mining techniques to improve the structure and
design of web pages/sites. As implemented today,
traversal path analysis, click-stream analysis,
optimal page space/ catalogue design, web-site
design/redesign, site activity mapping all qualify
under structure mining.

The site (or network) topology (neighbor pages,
leaf nodes, entry points, static versus dynamically
created pages, pages requiring user interaction,
etc.) may be input into structure mining. Further,
there are a few tools that get into the visualization/
mapping of this activity (Accrue, .Com Group, WUM,
AVS, MineIT and Angoss to name a few commercial
tools and research prototypes...)

Some useful references are:

Masand B. and Spiliopoulou M. WebKDD 99: Web Usage
Analysis and User Profiling. Proceedings of KDD-99
Workshop. http://www.acm.org/sigkdd/proceedings/webkdd99.
August, 99.

Buchner, A. et al. "Discovering Internet Marketing
Intelligence through Online Analytical Web Usage Mining,"
ACM SIGMOD Proceedings, April, 1998.

Cooley, R. et al. "Web Mining: Information and Pattern
Discovery on the World Wide Web." Proceedings of ICTAI
1997, Newport Beach, California. November 1997.

Cooley, R. et al. "Grouping Web Page References into
Transactions for Mining World Wide Web Browsing Patterns."
Proceedings of KDEX 1997. Newport Beach, California.
November 1997.

Pirolli, P. et al. "Silk from a sow's ear: Extracting
usable structures from the Web." Proc. of 1996 Conference
on Human Factors in Computing Systems (CHI-96), Vancouver,
British Columbia, Canada, 1996.

Pitkow, J. "In search of reliable usage data on the WWW."
Sixth International World Wide Web Conference, pages 451-463,
Santa Clara, CA, 1997.

Hope this is a good starter for you.

Regards,

      Ismail

Ismail Parsa
Epsilon
50 Cambridge Street
Burlington MA 01803 USA
781.685.6734 (tel)
781.685.0806 (fax)





Pablo Ibarra Casals <pibarra@tilsor.com.uy>@nautilus-sys.com on 04/06/2000
10:35:48 AM

Please respond to datamine-l@nautilus-sys.com

Sent by:  owner-datamine-l@nautilus-sys.com


To:   Adrķan Inciura <ainciura@tilsor.com.uy>
cc:   Datamining list <datamine-l@nautilus-sys.com>

Subject:  Re: DM: Datamining in Internet...



Hello,
          Does anyone knows something about datamining in the world wide web
architecture?, the idea is download information (maybe structural
information) about a knowledge domain (i.e. literature) and apply technics
to this data.
Many people tell me about Webmining, but all information I founded in this
topic is about web logs and customers tends. I don't want apply data mining
just in the content of web pages, I want apply this ones in the
architecture, viewing the net like a graph.

I would like to know any experiences or results on this matter.

Thanks a lot in advance, the suggestions are welcome.
Sincerely,
Pablo.

• Prev by Date: DM: CIA-2000 Call for Participation
• Next by Date: DM: Data Mining Tools comparison
• Prev by thread: Re: DM: Datamining in Internet...
• Next by thread: DM: clickstream Sequence Analysis
• Index(es):
  • Date
  • Thread