![]() |
|
![]() |
![]() |
|
![]() |
![]() |
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Subscribe]
Re: DM: Medical data mining: successful examples?From: Kamal M Ali Date: Wed, 25 Mar 1998 15:22:36 -0500 (EST)
Hello Rao,
IBM has done work in applying data-mining to patient
records. Specifically the project used association rules to
look for medical tests that could be predicted from
combinations of other medical tests. We represented
data/information about the patient as well as his/her medical
history in association rule format rather than the ususal
fixed col format cuz this is a variable information length
problem which cannot usefully be converted into the usual
fixed-lenght type of formats used by conventional
classification/predeiction tools. There is also a paper on
this in KDD 97 by myself, Srikant Ramakrishnan and Stefanos
Manganaris.
Kamal.
================================================================
Kamal Mahmood Ali, Ph.D. Phone: 408
927 1354
Senior data mining analyst, Fax: 408
927 3012
Global Business Intelligence Solutions, IBM Office: ARC
D3-250
IBM: http://direct.boulder.ibm.com/bi/tech/mining
IBM/GBIS Internal web pg: http://datamining.ibmus2.ibm.com
Personal web page: http://www.ics.uci.edu/~ali
================================================================
|
MHonArc 2.2.0