Papers about the state-of-the-art
State-of-the-art summarisation:
| Extraction task* | |||
| System | precision | recall | F-measure |
| Edinburgh | 85% | 82% | 83.47% |
| HeidelTime-1 | 90% | 82% | 85.81% |
| HeidelTime-2 | 82% | 91% | 86.27% |
| TipSem | 92% | 80% | 85.58% |
| TRIOS/TRIPS | 85% | 85% | 85.00% |
| Normalisation task* | ||
| System | value accuracy | type accuracy |
| Michele's normaliser | 86% | 95% |
| HeidelTime | 85% | 96% |
| TRIPS/TRIOS | 76% | 94% |
| Edinburgh-LTG | 63% | 84% |
* All the scores have been computed with respect to TempEval-2 test set.