D08-1111 . In our method , we employ a machine-learning method to train features ' weights
C04-1131 . Our approach uses supervised machine-learning techniques to automatically acquire
C00-1055 have enough raw data to begin machine-learning a way to distinguish these kinds
D09-1137 the corresponding document . The machine-learning model is constructed automatically
D08-1100 dialog analysis process through a machine-learning ap - proach . By inferring the
C02-1025 sequence of many hand-coded rules and machine-learning modules . 6 Conclusion We have
D09-1129 features . Another example of a machine-learning method is that of Carlson et
D08-1100 potentially be identified through a machine-learning approach . When comparing among
C04-1186 in developing a deterministic machine-learning based approach for dependency
D08-1111 segmentation task . To our knowledge , machine-learning methods used in segmentation
D09-1096 efforts to apply more sophisticated machine-learning techniques to identifying zone
D09-1101 ) ) . Bridging the gap between machine-learning approaches and linguistically-motivated
C00-1082 Because it ; was not clear which machine-learning method would 1 ) e the one most
D08-1108 our work is the first published machine-learning approach to productively model
C00-1055 preliminary , positive results in machine-learning the difference between human-produced
C00-1082 mnnber of man - hours , we used machine-learning methods for bunsetsu identitication
D08-1100 interesting to see how well a machine-learning approach can perform on the problem
D08-1111 train features ' weights . Many machine-learning methods , such as HMM ( Zhang
D09-1062 either a heuristic-based one , or a machine-learning based one -- we consider it as
D08-1111 according to this rank . Other machine-learning based segmentation algorithms
hide detail