This paper separates conditional parameter estima- tion, which consistently raises test set accuracy on statistical NLP tasks, from conditional model struc- tures, such as the conditional Markov model used for maximum-entropy tagging, which tend to lower accuracy. Error analysis on part-of-speech tagging shows that the actual tagging errors made by the conditionally structured model derive not only from label bias, but also from other ways in which the in- dependence assumptions of the conditional model structure are unsuited to linguistic sequences. The paper presents new word-sense disambiguation and POS tagging experiments, and integrates apparently conflicting reports from other recent work.

Click Here to View the Paper In XML Format