I found John Cuadrado's "Mining Statistics" (February) very interesting. My students and I are working on a problem for the Navy related to the collection of medical symptoms data from military personnel. The Navy is aware of algorithms and expert systems that take this data and output a probable diagnosis. What they want is a method of aggregating the data sequentially (as it comes in on a patient-by-patient basis) into a few continuously monitored measurements. They want to identify increasing patterns of illness quickly.
My idea is to use the geographical information on endemic diseases to establish known centroids in "symptom" space. The hard part is that a significant number of patients that cannot be classified into one of the endemic diseases might accumulate. We need an algorithm that detects when a new cluster is required and t
hen adjusts the classification scheme accordingly.
John Angus
Professor of Mathematics
The Claremont Graduate School
Claremont, CA
angusj@cgs.edu or angusjohne@aol.com
Flexible C++
Matthew Wilson
My approach to software engineering is far more pragmatic than it
is
theoretical--and no language better exemplifies this than C++.
BYTE Digest editors every month analyze and evaluate the best articles from Information Week, EE Times, Dr. Dobb's Journal, Network Computing, Sys Admin,
and dozens of other CMP publications—bringing
you critical news and information about wireless communication,
computer security, software development, embedded systems,
and more!
RFA (Request for Algorithm)
