| W02-1008 | of this minimalist approach to | negative sample selection | . Evaluation . We evaluate the |
| W12-6107 | a positive sample and its two | negative sample selections | . There are 10 types of features |
| W12-6107 | . This section introduces a " | negative sample selection | " process , which fulfills the |
| W02-1008 | system and examine the effect of | negative sample selection | . Section 4 presents our corpus-based |
| W02-1008 | skewed class distributions -- | negative sample selection | , i.e. the selection of a smaller |
| W02-1008 | in the absence and presence of | negative sample selection | . Without negative sample selection |
| W02-1008 | is selected for NP ✟ . 3 | Negative Sample Selection | As noted above , skewed class |
| W02-1008 | negative sample selection . Without | negative sample selection | , F-measure increases from 52.4 |
| W12-6107 | extension . 3.3.1 Positive and | Negative Sample Selection | . In this term extraction process |
| W02-1008 | incorporating NEG - SELECT . With | negative sample selection | , the percentage of positive |
| W02-1008 | problems , we presented a minimalist | negative sample selection | algorithm to reduce the skewness |
| W02-1008 | for MUC-7 ) . Similarly , with | negative sample selection | , F-measure increases from 55.2 |
| W02-1008 | number of negative instances ( via | negative sample selection | ) improves recall but damages |
| W02-1008 | evaluate the coreference system with | negative sample selection | on the MUC-6 and MUC-7 coreference |
| W02-1008 | rules . We propose a method for | negative sample selection | that , for each anaphoric NP |
