Is there any medical data set downloadable for
research(statistics/data mining) purpose ?
For example, I can do some statistics on the data set to analyze the
correlation between some kind of diseases outbreaking in some area for
prediction/prevention purpose or even for biodefense research or
bioterrorism monitoring?
The data set should be in numerical format and not cancer data ('cause
there are lots of cancer data sets online). It is better that the data
set has lots of attributes(better more than 20 attributes) and
observations(better more than 100). |
Request for Question Clarification by
tar_heel_v-ga
on
17 Feb 2003 10:18 PST
olleh, elgoog_elgoog :)
Your request is a bit vague in the type of data you are looking for.
Would something that has a description as this suffice:
"..contain tabular and graphic information about reported cases
collected from 59 reporting areas (the 50 states, the District of
Columbia, New York City, U.S. dependencies and possessions, and
independent nations in free association with the United States). The
reports include statistics on case counts and case rates by states and
metropolitan statistical areas with tables of selected demographic and
clinical characteristics (e.g., race/ethnicity, age group, country of
origin, form of disease, drug resistance, etc"
-THV
|
Clarification of Question by
elgoog_elgoog-ga
on
17 Feb 2003 11:48 PST
Hi tar_heel_v,
Actually, I need some kind of medical data to phrase an application
scenario for my research(correlation statistics/data mining etc.).
So as long as the data you provided can be fitted in some real
application
scenario, it works.
However, the data should be in numerical format(at least most of the
attributes should be) so that I can do statistics on them directly.
It should not be statistics report but raw data.
The data you described looks like a statistics report. If not, it may
suffice my
requirements.
Thank you very much.
|
Request for Question Clarification by
tar_heel_v-ga
on
17 Feb 2003 19:07 PST
I am still a bit confused as to exactly what you are looking for,
therefore, here is the link to the data:
http://www.cdc.gov/nchstp/tb/surv/surv2001/default.htm
If this meets the criteria that you are looking for, let me know and I
will post as final answer as well as one or two additional sources
with similar data.
-THV
|
Clarification of Question by
elgoog_elgoog-ga
on
18 Feb 2003 00:28 PST
Hi tar_heel_v,
Thank you very much for your kind help.
I took a look at that data set. I am sorry but it is not
the data I am looking for. It is more like a statistical survey than
raw medical record.
Here is an example I prefer--
http://lisp.vse.cz/pkdd99/Challenge/tsumoto.htm
I am looking for this kind of data set. However, this data set has
lots of
missing values(it is reasonable for real data mining application, but
this causes a little trouble for our current algorithm.)
I appreciate if you could find some data sets as the example without
missing values. If it is difficult to find one, that is OK.
Thank you very much.
|