Hypernym Discovery On Medical Domain
Metrics
MAP
MRR
P@5
Results
Performance results of various models on this benchmark
Comparison Table
| Model Name | MAP | MRR | P@5 |
|---|---|---|---|
| supervised-distributional-hypernym-discovery | 18.84 | 41.07 | 20.71 |
| crim-at-semeval-2018-task-9-a-hybrid-approach | 34.05 | 54.64 | 36.77 |
| 300-sparsans-at-semeval-2018-task-9-hypernymy | 20.75 | 40.60 | 21.43 |
| expr-at-semeval-2018-task-9-a-combined | 13.77 | 40.76 | 12.76 |
| adapt-at-semeval-2018-task-9-skip-gram-word | 8.13 | 20.56 | 8.32 |
| sjtu-nlp-at-semeval-2018-task-9-neural | 11.69 | 25.95 | 11.69 |
| Model 7 | 28.93 | 35.80 | 34.20 |
| hypernyms-under-siege-linguistically | 0.91 | 2.10 | 1.08 |