Title :
On functional relation between class-selective rejection error and average number of classes
Author_Institution :
Inst. of Comput. Sci. & Appl. Math., Bern Univ., Switzerland
Abstract :
The paper first reviews the recently proposed optimum class-selective rejection rule. This rule provides an optimum trade-off between the error rate and the average number of (selected) classes. Then, a new general relation between the error rate and the average number of classes is presented. The error rate can directly be computed from the class-selective reject function, which in turn can be estimated from unlabelled patterns, by simply counting the rejected classes. Theoretical as well as practical implications are discussed and some future research directions are proposed
Keywords :
error analysis; error statistics; estimation theory; optimisation; pattern classification; probability; average class number; class-selective rejection error; error rate; functional relation; optimisation; pattern classification; probability; unlabelled patterns; Application software; Computer applications; Computer errors; Computer science; Error analysis; Inspection; Mathematics; Pattern recognition; Probability density function; System performance;
Conference_Titel :
Intelligence and Systems, 1996., IEEE International Joint Symposia on
Conference_Location :
Rockville, MD
Print_ISBN :
0-8186-7728-7
DOI :
10.1109/IJSIS.1996.565080
