Effect of Separate Sampling on Classification and the Minimax Criterion Conference Paper

abstract

• It is commonplace in bioinformatics (and elsewhere) to build a classifier from sample data in which the sample sizes of the classes are not random; that is, they are selected prior to sampling. The result is that there is no estimate of the prior class probabilities available from the data. In this paper, we find an analytic result for the minimax solution for the class prior probabilities for a general Neyman-Pearson induced classifier. From that we derive Anderson's classical minimax prior probability 'estimate.' Using synthetic and real data, we demonstrate the degradation in classifier performance from using inaccurate values for the prior probabilities. 2013 IEEE.

name of conference

• 2013 IEEE International Workshop on Genomic Signal Processing and Statistics

authors

• Dougherty, Edward

published proceedings

• 2013 IEEE INTERNATIONAL WORKSHOP ON GENOMIC SIGNAL PROCESSING AND STATISTICS (GENSIPS 2013)

author list (cited authors)

• Esfahani, M. S., & Dougherty, E. R.

citation count

• 1

complete list of authors

• Esfahani, Mohammad Shahrokh||Dougherty, Edward R

publication date

• November 2013

publisher

• Institute of Electrical and Electronics Engineers (IEEE)  Publisher

published in

• IEEE International Workshop on Genomic Signal Processing and Statistics : [proceedings]. IEEE International Workshop on Genomic Signal Processing and Statistics  Journal

keywords

• Classification Accuracy
• Minimax
• Separate Sampling

Digital Object Identifier (DOI)

• 10.1109/gensips.2013.6735935

International Standard Book Number (ISBN) 13

• 9781479934621

start page

• 72

end page

• 73

URL

• http%3A%2F%2Fdx.doi.org%2F10.1109%2Fgensips.2013.6735935