| W02-2028 | NB classification combined with | greedy clustering | . In the case of greedy clustering |
| P08-1096 | The resolution is done using a | greedy clustering | strategy . Given a test document |
| S10-1022 | of the classifi - er , using a | greedy clustering | algorithm . Each mention is compared |
| W12-4511 | incorporated as constraints during | greedy clustering | . For Chinese , we used relations |
| W12-4511 | spectral clustering before the final | greedy clustering | phase . In order to be able to |
| S12-1002 | coreference resolution as the | greedy clustering | process shown in Algorithm 1 |
| S12-1002 | shown in Algorithm 3 . Like the | greedy clustering | of Algorithm 1 , it starts with |
| P08-1096 | For simplicity , we just use a | greedy clustering | strategy for resolution , that |
| W12-4511 | mentions . Entities are obtained via | greedy clustering | . We participated in the closed |
| N06-1046 | on grouping entry pairs over a | greedy clustering | - based model which does not |
| W12-4511 | position in the text and perform | greedy clustering | ( Section 2.6 ) . For Chi - nese |
| S10-1020 | coreferent or not . During testing , a | greedy clustering | algorithm ( link-first ) is next |
| W12-4511 | Cai et al. ( 2011b ) before the | greedy clustering | step to reduce the number of |
| W02-2028 | n't include the results of the | greedy clustering | into Figure 2 . In Table 2 , |
| W12-4511 | computation , graph construction and | greedy clustering | look at all pairs of mentions |
| P13-3012 | Mitkov , 1998 ) can be regarded as | greedy clustering | in a multigraph , where edges |
| P05-1020 | clustered , employing instead a | greedy clustering | procedure to construct a partition |
| P11-1080 | into entities with some form of | greedy clustering | using a pairwise mention similarity |
| W01-0717 | is that the first system uses | greedy clustering | while COR , UDIS optimizes using |
| W02-2028 | greedy clustering . In the case of | greedy clustering | , it is necessary to display |
