Hello,
We've generated software components that can data mine data in XML. A
moments thought will permit you to see that this is not as trivial as it
might seem. Because Xml represents tree structures it offers a real
challenge to datamining.
1) Relational data is "orthogonal" i.e. representable in rows and columns
with no gaps. XML data isn't, or needn't be to be valid XML. If you try to
convert it to row column format it can be very sparse, and thus Neural nets
etc will fail to find useful relationships. Special algorithms are required.
2) There's more to mine in XML. Depending on the schema, nodes and subtrees
may be optional, and inferences could be made from the absence or presence
of subtrees.
3) XML trees can encode relationships that can also be mined, for instance a
family tree might be encoded in XML, and inferences might be made about the
life expectancy of an individual depending on his/her parents. The
opportunities to mine relationships are vast, and an untapped area for data
mining.
We represent the knowledge discovered in an XML based schema called
Metarule, in the form of fuzzy logic expert system rules. We have software
objects that can read this knowledge and apply it to XML documents,
inserting predictions/classifications as required. The opportunity exists to
create
your own expert systems that act on XML directly, and supply a guided editor
for this purpose.
If you think about the above, I think you can see that this is a technology
with a lot of potential. Our problem is finding applications that prove the
usefulness of this technology now. We are working on a few examples but
would like to hear from academic or commercial organisations that have a
data set that challenges our technology and that has interesting
applications.
We're a commercial organization, but would be happy to waive fees and
charges for the right application up to commercial exploitation.
I hope you can help,
Andrew N. Edmonds Ph.D. MBCS
President
Scientio, inc.
Haydon House
Station Road
Woburn Sands
Bucks MK17 8RX UK
tel: +44 1908-584226
fax: +44 1908-584236
email: andy@metadatamining.com
web: www.metarule.com
############################################################################
This message was posted through the fuzzy mailing list.
(1) To subscribe to this mailing list, send a message body of
"SUB FUZZY-MAIL myFirstName mySurname" to listproc@dbai.tuwien.ac.at
(2) To unsubscribe from this mailing list, send a message body of
"UNSUB FUZZY-MAIL" or "UNSUB FUZZY-MAIL yoursubscription@email.address.com"
to listproc@dbai.tuwien.ac.at
(3) To reach the human who maintains the list, send mail to
fuzzy-owner@dbai.tuwien.ac.at
(4) WWW access and other information on Fuzzy Sets and Logic see
http://www.dbai.tuwien.ac.at/ftp/mlowner/fuzzy-mail.info
(5) WWW archive: http://www.dbai.tuwien.ac.at/marchives/fuzzy-mail/index.html
This archive was generated by hypermail 2b30 : Fri Mar 29 2002 - 16:01:44 MET