W06-1672 superior to local classifier-based discriminative modeling . This may have resulted from
W06-1672 probabilistic modeling , with the global discriminative modeling approach achieving the best performance
W06-1672 name transliteration , global discriminative modeling is superior to local classifier-based
W09-3526 features . Table 2 demonstrates that discriminative modeling significantly improves performance
J10-3002 Feature Functions The primary art in discriminative modeling is to define useful features
W10-1763 , using various generative and discriminative modeling techniques . For example , Ananthakrishnan
W06-0125 of contexts and implement the discriminative modeling in MIIM . The third step is post
W06-1672 all languages , with the global discriminative modeling approach achieving the best performance
W09-3526 In general , MDL training with discriminative modeling allows us to discover a flexible
W09-3526 transliteration and combines it with discriminative modeling . We apply the proposed approach
J10-3005 features and incorporated into discriminative modeling paradigms ( e.g. , Nguyen et
P10-1147 entities and numbers . <title> Discriminative Modeling of Extraction Sets for Machine
D09-1123 dependency parsing mechanism using the discriminative modeling capabilities . Acknowledgments
P11-1042 there has been no previous work on discriminative modeling of Urdu , since , to our knowledge
P11-1042 ever , unlike previous work on discriminative modeling of word alignment ( which also
W09-3526 Transliteration Training with Discriminative Modeling Dmitry </title> <authors></authors>
W14-0125 used TADM ( Toolkit for Advanced Discriminative Modeling ; Malouf , 2002 ) for the training
P06-1026 richer feature set we use and a discriminative modeling framework that supports a large
W12-3021 because they provide a framework for discriminative modeling while succinctly representing
N06-1036 Sutton et al. , 2004 ) and general discriminative modeling on structured outputs ( Bartlett
hide detail