Table 1 Evaluation results. Counts of positive and negative examples in the corpora and (P)recision, (R)ecall, (F)-score and AUC for the graph kernel, with standard deviations provided for F and AUC.
Statistics | Graph Kernel | co-occ | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
Corpus | #POS. | #NEG. | P | R | F | σ F | AUC | σ AUC | P | F | ||
AIMed | 1000 | 4834 | 52.9% | 61.8% | 56.4% | 5.0% | 84.8% | 2.3% | 17.8% | 30.1% | ||
BioInfer | 2534 | 7132 | 56.7% | 67.2% | 61.3% | 5.2% | 81.9% | 4.9% | 26.6% | 41.7% | ||
HPRD50 | 163 | 270 | 64.3% | 65.8% | 63.4% | 11.4% | 79.7% | 6.3% | 38.9% | 55.4% | ||
IEPA | 335 | 482 | 69.6% | 82.7% | 75.1% | 7.0% | 85.1% | 5.1% | 40.8% | 57.6% | ||
LLL | 164 | 166 | 72.5% | 87.2% | 76.8% | 17.8% | 83.4% | 12.2% | 55.9% | 70.3% | ||