| W06-1710 | - ror . Further , we perform a | binary categorization | for each of the 16 categories |
| M91-1035 | generating the templates from the | binary categorization | . Since the categorizer output |
| W13-4049 | < 2.5 average ) . With this | binary categorization | , 44.8 % of the time tutor responses |
| P14-2118 | create a distinct but analogous | binary categorization | problem to the composition task |
| W11-1516 | success rate of 62.5 % in its | binary categorization | of collaborations . 5.3.2 First |
| W15-0616 | revisions using Faigley and Witte 's | binary categorization | , only the number of textbased |
| H92-1041 | used in our experiments . 3.1 . | Binary Categorization | In order to compare text categorization |
| W98-1508 | , it is possible to consider a | binary categorization | problem where a decision is made |
| W98-1108 | , it is possible to consider a | binary categorization | problem where a decision is made |
| M91-1035 | generated a template file from the | binary categorization | and evaluated the resulting categorization |
| M91-1035 | use to generate templates from a | binary categorization | . Many of the slot filling tasks |
| W10-3021 | Although the problem itself is a | binary categorization | problem , we approach the problem |
| P15-1144 | Newsgroup Dataset . We consider three | binary categorization | tasks from the 20 Newsgroups |
| M91-1035 | roughly 50/50 when evaluated under a | binary categorization | model and around 35/35 under |
| W01-0508 | documents . Reuters-21578 is a | binary categorization | corpus , meaning that documents |
| P14-1145 | to measure ( than the coarser | binary categorization | of polarities ) , our connotation |
| M91-1035 | categorizer , and the correct | binary categorization | decisions extracted fro m the |
| M91-1035 | Since the categorizer output only | binary categorization | decisions , cross references |
| W01-0508 | accuracy ( or overall er - ror ) for | binary categorization | , since most documents have few |
| C04-1195 | Perhaps what is needed then is not a | binary categorization | but a continuous one , a numeric |
