Index Catalog // ScholarsArchive@OSU

Start Over

Filtering by: Creator Johnson, Colin P. Degree Level Doctoral Resource Type Technical Report

Model Selection of Discrete Classifiers under Class and Cost Distribution Change: An Empirical Study

Creator:

Ahmed, Nahian and Hutchinson, Rebecca A.

Abstract:

A variety of important machine learning applications require predictions on test data with different characteristics than the data on which a model was trained and validated. In particular, test data may have a different relative frequency of positives and negatives (i.e., class distribution) and/or different mislabeling costs of false positive...

Resource Type:

Technical Report

Full Text:

performance (Provost & Fawcett, 1997). To see this, consider the expected cost of a classifier P (Y = 1) · (1