Send to

Choose Destination
IEEE Trans Pattern Anal Mach Intell. 2012 Feb;34(2):292-301. doi: 10.1109/TPAMI.2011.163.

Does the cost function matter in Bayes decision rule?

Author information

Lehrstuhl für Informatik 6, Computer Science Department, RWTH Aachen University, Ahornstr. 55, Aachen 52074, Germany.


In many tasks in pattern recognition, such as automatic speech recognition (ASR), optical character recognition (OCR), part-of-speech (POS) tagging, and other string recognition tasks, we are faced with a well-known inconsistency: The Bayes decision rule is usually used to minimize string (symbol sequence) error, whereas, in practice, we want to minimize symbol (word, character, tag, etc.) error. When comparing different recognition systems, we do indeed use symbol error rate as an evaluation measure. The topic of this work is to analyze the relation between string (i.e., 0-1) and symbol error (i.e., metric, integer valued) cost functions in the Bayes decision rule, for which fundamental analytic results are derived. Simple conditions are derived for which the Bayes decision rule with integer-valued metric cost function and with 0-1 cost gives the same decisions or leads to classes with limited cost. The corresponding conditions can be tested with complexity linear in the number of classes. The results obtained do not make any assumption w.r.t. the structure of the underlying distributions or the classification problem. Nevertheless, the general analytic results are analyzed via simulations of string recognition problems with Levenshtein (edit) distance cost function. The results support earlier findings that considerable improvements are to be expected when initial error rates are high.

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for IEEE Engineering in Medicine and Biology Society
Loading ...
Support Center