Error Measures and Bayes Decision Rules Revisited with Applications to POS Tagging
Hermann Ney, Maja Popović, D. Heinr. Sundermann · 2004
Starting from first principles, we re-visit the statisti-cal approach and study two forms of the Bayes deci-sion rule: the common rule for minimizing the num-ber of string errors and a novel rule for minimizing the number of symbols errors. The Bayes decision rule for minimizing the number of string errors is widely used, e.g. in speech recognition, POS tag-ging and machine translation, but its justification is rarely questioned. To minimize the number of sym-bol errors as is more suitable for a task like POS tag-ging, we show that another form of the Bayes deci-sion rule can be derived. The major purpose of this paper is to show that the form of the Bayes decision rule should not be taken for granted (as it is done in virtually all statistical NLP work), but should be adapted to the error measure being used. We present first experimental results for POS tagging tasks. 1