Skip to main content

Table 5 Performance of submitted classifiers on test data.

From: A linear classifier based on entity recognition tools and a statistical approach to method extraction in the protein-protein interaction literature

Classifier Features F 1 Accuracy MCC AUCiP/R
VTT0 SP 0.5399 0.8097 0.456 0.4935
VTT0 Bigrams 0.5243 0.8382 0.4318 0.4287
VTT1 SP 0.5667 0.8213 0.4909 0.5402
VTT1 Bigrams 0.5575 0.8402 0.472 0.5015
VTT5 SP 0.6483 0.864 0.5897 0.7339
VTT5 Bigrams 0.6366 0.85.9 0.5752 0.7127
VTT3 SP 0.628 0.8387 0.5735 0.7143
  1. Shown are the values obtained on the official BC3 gold standard by the F-Score, Accuracy, Matthew’s Correlation Coefficient, and Area Under the interpolated Precision and Recall Curve (computed with the official script, and adding F-Score). Boldfaced values represent best performance in table.