Skip to main content
11 events
when toggle format what by license comment
Mar 27, 2012 at 7:49 comment added Jim Bo Actually yes, currently the diagnosis is a composite score, along with biopsies & other measures. However I intend to build the classifier without using the tests and just using the biopsy/other measures to determine the class label, to avoid circularity
Mar 27, 2012 at 5:30 answer added SheldonCooper timeline score: 0
Mar 26, 2012 at 23:38 history tweeted twitter.com/#!/StackStats/status/184424219927515137
Mar 26, 2012 at 18:21 comment added Matt Parker Were any of the 15 tests used to establish your current disease diagnoses, or was there a separate (and hopefully great) standard test that determined them?
Mar 26, 2012 at 18:07 answer added MSPAN timeline score: 1
Mar 26, 2012 at 18:04 comment added Jim Bo Sorry if it was unclear! I thought the fact I could see a couple of clusters in the data might be helpful in the choise of model/classifier/rule
Mar 26, 2012 at 17:49 comment added Jim Bo Not necessarily from all features.. it would also be great if we could find a sub-set that is as good as using all 15.
Mar 26, 2012 at 17:46 comment added Jim Bo Exactly. I did the clustering to look for "structure" in the data, i.e. if I could see difference in patterns of scores between dis vs. non-disease. However my aim is to devise a (ideally simple and intuitive) scoring rule
Mar 26, 2012 at 17:43 comment added chl It seems you want to use some kind of a supervised algorithm (where class membership is known in advance) as opposed to unsupervised methods (like clustering). Can you confirm that what you are after is simply devising a 'scoring rule' (Prob(individual i has disease)) from all 15 features?
Mar 26, 2012 at 17:40 history edited chl CC BY-SA 3.0
added 2 characters in body
Mar 26, 2012 at 15:40 history asked Jim Bo CC BY-SA 3.0