OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help



   RE: [xml-dev] Data mining the semantic web? (was RE: [xml-dev] Semantic

[ Lists Home | Date Index | Thread Index ]

As long as you remember:

1.  Ontologies have to be shared; interpretations can still vary.

2.  Information has to be shared.  What you can't see, you can't mine.

3.  A stratified web will return uneven results for given mining efforts.

4.  You have to filter for superstition.  A system that cannot 
    distinguish between identification and classification is not one 
    on which to build quick decision systems.  Metaphorically, don't 
    build fire control systems over URIs.

5.  One wants to combine "opinions of others" with "self-published 
    opinions".  Prediction also relies on self-asserted opinions.

6.  All a data mining effort can return is an informed opinion.

Yeah, DARPA wants it bad.  Has since before there was a web to mine.


-----Original Message-----
From: Joshua Allen [mailto:joshuaa@microsoft.com]

Completely in agreement with Jonathan here.  This intersection is
interesting to many people it seems: http://www.darpa.mil/iao/ 


News | XML in Industry | Calendar | XML Registry
Marketplace | Resources | MyXML.org | Sponsors | Privacy Statement

Copyright 2001 XML.org. This site is hosted by OASIS