Share Email Print
cover

Proceedings Paper

A risk-based comparison of classification systems
Author(s): Steven Thorsen; Seth Wagenman; Mark E. Oxley; Kenneth W. Bauer
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Performancemeasures for families of classification system families that rely upon the analysis of receiver operating characteristics (ROCs), such as area under the ROC curve (AUC), often fail to fully address the issue of risk, especially for classification systems involving more than two classes. For the general case, we denote matrices of class prevalence, costs, and class-conditional probabilities, and assume costs are subjectively fixed, acceptable estimates for expected values of class-conditional probabilities exist, and mutual independence between a variable in one such matrix and those of any other matrix. The ROC Risk Functional (RRF), valid for any finite number of classes, has an associated parameter argument, which specifies a member of a family of classification systems, and for which there is an associated classification system minimizing Bayes risk over the family. We typify joint distributions for class prevalences over standard simplices by means of uniform and beta distributions, and create a family of classification systems using actual data, testing independence assumptions under two such class prevalence distributions. Examples are given where the risk is minimized under two different sets of costs.

Paper Details

Date Published: 11 May 2009
PDF: 12 pages
Proc. SPIE 7336, Signal Processing, Sensor Fusion, and Target Recognition XVIII, 73360T (11 May 2009); doi: 10.1117/12.819280
Show Author Affiliations
Steven Thorsen, U.S. Air Force Academy (United States)
Seth Wagenman, Air Force Institute of Technology (United States)
Mark E. Oxley, Air Force Institute of Technology (United States)
Kenneth W. Bauer, Air Force Institute of Technology (United States)


Published in SPIE Proceedings Vol. 7336:
Signal Processing, Sensor Fusion, and Target Recognition XVIII
Ivan Kadar, Editor(s)

© SPIE. Terms of Use
Back to Top