D09-1158 learning algorithm is required in bootstrapping approaches . We describe the two machine
C04-1151 propose an effective , multi-level bootstrapping approach to accomplish this task ( Figure
D09-1149 Therefore , it is critical for a bootstrapping approach to select the most appropriate
D13-1168 that could be exploited in the bootstrapping approach . In order to induce initial
D13-1168 that are used as seeds for the bootstrapping approach ( see sect . 2.2 ) , we experiment
D12-1129 inventories like WordNet and propose a bootstrapping approach to the acquisition of sense inventories
D10-1034 systematically evaluate our stratified bootstrapping approach in the semantic relation classification
C04-1078 Conclusion We have presented a novel bootstrapping approach for information extraction by
D12-1129 Riloff ( 2010 ) , who devised a bootstrapping approach to induce semantic class taggers
C04-1199 semantic relations , in effect a bootstrapping approach . Applying this process by hand
D09-1033 This is a well-known problem for bootstrapping approaches ( Blum and Mitchell , 1998 ;
D09-1158 domains . 6 Conclusion We proposed a bootstrapping approach for domain adaptation , and we
D10-1034 section systematically evaluates the bootstrapping approach using clustering-based stratified
D13-1168 the different components of the bootstrapping approach in the task of bilingual lexicon
D10-1034 clustering - based stratified bootstrapping approach achieves the best F1-score of
D13-1168 black box to acquire seeds for the bootstrapping approach , but we encourage the interested
D11-1132 scale for embedding . Compared to bootstrapping approach , our approach is accumulative
D13-1168 external translation resources , our bootstrapping approach yields lexicons that outperform
D12-1057 Japanese in Section 5.1 . 3.2 Bootstrapping Approach to Excitation Template Acquisition
D13-1018 domain of inter - est , we apply a bootstrapping approach which iteratively obtains generalized
hide detail