Advertisement
Table 4 Performance of different CWS tools on webMedQA with MV-LSTM
From: Applying deep matching networks to Chinese medical question answering: a study and a dataset
| Vocab Size | P@1(%) | MAP(%) | |
|---|---|---|---|
| Ansj | 44140 | 57.7 | 73.5 |
| Fnlp | 145058 | 57.9 | 74.4 |
| jieba | 94630 | 59.3 | 75.3 |
