| W13-4901 | Experimental Setup The main algorithm for | agreement-based co-training | is given in Figure 5 . Again |
| W13-4901 | selection algorithm we used for | agreement-based co-training | ( Figure 5 ) . For this ex - |
| W03-0407 | similar to those produced by our | agreement-based co-training | method . After 50 rounds of co-training |
| W03-0407 | literature . Our results show that | agreement-based co-training | can significantly improve tagging |
| W03-0407 | annotated data . It is possible that | agreement-based co-training | , using more careful selection |
| W13-4901 | the non-iterative 85 % threshold | agreement-based co-training | approach described in Section |
| W03-0407 | in the previous experiments , | agreement-based co-training | required the taggers to be re-trained |
| W11-3906 | Brown test set are consistent with | agreement-based co-training | . However , the separation of |
| W03-0407 | 500 or 50 seed sentences , and | agreement-based co-training | was applied , using a cache size |
| W03-0407 | co-training . There are advantages to | agreement-based co-training | , however . First , the agreement-based |
| W03-0407 | , yield comparable results to | agreement-based co-training | , with only a fraction of the |
| W13-4901 | also follow the algorithm for | agreement-based co-training | as presented in Figure 5 . However |
| W13-3502 | Blum and Mitchell , 1998 ) , and | agreement-based co-training | ( Clark et al. , 2003 ) in particular |
| W03-0403 | active learning as the inverse of | agreement-based co-training | ( Abney , 2002 ) . We can then |
| W11-3906 | to 0.7 and 0.9 point gains by | agreement-based co-training | with feature split BS shows that |
| W03-0407 | improvement of 73.2 % to 85.9 % . Our | agreement-based co-training | results support the theoretical |
| W03-0407 | The other experiment used our | agreement-based co-training | approach ( 50 seed sentences |
| W14-6105 | default settings ( see Table 10 ) . | Agreement-based co-training | . We first experimented with |
| W13-4901 | selection in co-training . 5.3.2 | Agreement-Based Co-Training | Experimental Setup The main algorithm |
| W03-0407 | 4.2 Naive Co-training Results | Agreement-based co-training | for POS taggers is effective |
