How to do classification with high predictive value?

[dchalmer]

I have a two-class classification problem. I would like to train a multivariate classifier with 100% positive predictive value. In other words, I want the model to completely avoid one of the classes. For this application a low-ish sensitivity is OK as long as PPV is ~100%.
Do you have any suggestions of good techniques to use?

Thank you!

[aridneptune]

This isn't the right forum for this sort of question. Try Stackexchange.http://stats.stackexchange.com/

[Planet]

I have a two-class classification problem. I would like to train a multivariate classifier with 100% positive predictive value. In other words, I want the model to completely avoid one of the classes. For this application a low-ish sensitivity is OK as long as PPV is ~100%.

Do you have any suggestions of good techniques to use?

Thank you!

 

 

I know an article that did precisely what you ask. 

 

Dr. Harvey wanted to predict deep vein thrombosis (DVT) using d-dimer (DD) as an attribute, but he sought 100% correct prediction of all positive DVD cases.  So he ordered observations by their d-d value, and located the d-d value beneath which there were no observations positive for DVD.  He then assessed the statistical significance and effect strength of the resulting model using an exact non-parametric methodology known as “optimal data analysis” (ODA).  The ODA statistical paradigm explicitly maximizes model accuracy (rather than variance or value of the likelihood function).  Dr. Harvey’s article was selected by the American College of Physicians Journal Club, and read by most practicing internists in the US.

 

Here is the link to the article citation: http://stroke.ahajournals.org/content/27/9/1516.long

 

Here is the citation to the seminal introduction to the ODA paradigm, which comes with software for Windows: http://www.amazon.com/Optimal-Data-Analysis-Guidebook-Software/dp/1557989818

 

And here is a blog about ODA: http://odajournal.wordpress.com/

 

BTW, regardless of the analytic geometry (the structure of the data), ODA always identifies the model that explicitly maximizes model classification accuracy. 

Edited September 30, 2013 by Planet