| D08-1105 | maximize the benefits by performing | active learning | only on the more frequently occurring |
| D08-1105 | using feature augmentation with | active learning | . Our results show that this |
| C04-1201 | is a very time consuming task , | active learning | can provide a faster approach |
| D08-1105 | unlikely event 1007 Figure 2 : The | active learning | algorithm . that we have access |
| C04-1201 | future work we plan to investigate | active learning | with SVM for this problem . Given |
| D08-1112 | comments . <title> An Analysis of | Active Learning | Strategies for Sequence Labeling |
| D08-1105 | various curves stabilize after 35 | active learning | iterations , we only show the |
| D08-1105 | domain adaptation technique with | active learning | , we are able to effectively |
| D08-1105 | our adaptation exam ples during | active learning | . Hence , we perform active learning |
| D08-1105 | curves the results of applying | active learning | only to various sets of word |
| D08-1105 | types . In contrast , we perform | active learning | experiments on the hundreds of |
| D08-1112 | </title> <authors></authors> Abstract | Active learning | is well-suited to many problems |
| D08-1105 | by the OntoNotes data . For our | active learning | experiments , we use the uncertainty |
| D08-1105 | examples that have been selected via | active learning | thus far . We then use the AUGMENT |
| D08-1105 | introduced by Daume III ( 2007 ) , and | active learning | ( Lewis and Gale , 1994 ) to |
| D08-1112 | aims to shed light on the best | active learning | approaches for sequence labeling |
| D08-1105 | examples to annotate , we could use | active learning | ( Lewis and Gale , 1994 ) to |
| D08-1105 | WSD accuracy of 82.6 % after 10 | active learning | iterations . Note that in Section |
| D08-1105 | technique during each iteration of | active learning | to combine the SEMCOR examples |
| D08-1105 | select examples to annotate via | active learning | . Also , since we have found |
