Table 4 Evaluation results of the approach on cross-corpus datasets (we report percentages); Here 'FM' denotes 'F-measure'
From: Biomedical named entity extraction: some issues of corpus compatibilities
Approach | Training set | Test set | Recall | Precision | FM |
|---|---|---|---|---|---|
Best Ind. Classifier | JNLPBA (protein only)+AIMed | AIMed | 83.14 | 83.19 | 83.17 |
SOO | JNLPBA (protein only)+AIMed | AIMed | 85.10 | 85.01 | 85.05 |
Best Ind. Classi | JNLPBA (protein + DNA)+AIMed | AIMed | 82.17 | 84.15 | 83.15 |
SOO | JNLPBA (protein + DNA)+AIMed | 3-fold cross | 84.07 | 86.01 | 85.03 |
validation on AIMed | |||||
Best Ind. Classi | JNLPBA (protein only)+GENETAG | GENETAG | 89.44 | 93.07 | 91.22 |
SOO | JNLPBA (protein only)+GENETAG | GENETAG | 91.19 | 94.98 | 93.05 |
Best Ind. Classi | JNLPBA (protein + DNA + RNA)+GENTAG | GENTAG | 88.70 | 93.55 | 91.06 |
SOO | JNLPBA (protein + DNA + RNA)+GENTAG | GENTAG | 90.09 | 95.16 | 92.56 |