Skip to main content

Table 5 Performance on the test set (95% CI = 95% confidence interval)

From: C-Norm: a neural approach to few-shot entity normalization

  Habitats Phenotypes
  Wang [95% CI] Strict [95% CI] Wang [95% CI] Strict [95% CI]
Baseline 0.559 [0.543, 0.576] 0.224 [0.199, 0.250] 0.581 [0.559, 0.604] 0.091 [0.056, 0.127]
AmritaCen 0.522 [0.497, 0.548] 0.347 [0.314, 0.376] 0.646 [0.595, 0.698] 0.512 [0.448, 0.571]
BOUN-ISIK 0.687 [0.667, 0.710] 0.428 [0.395, 0.459] 0.566 [0.520, 0.610] 0.315 [0.259, 0.373]
BLAIR GMU 0.615 [0.596, 0.632] 0.211 [0.185, 0.237] 0.646 [0.607, 0.685] 0.313 [0.254, 0.373]
PADIA 0.684 [0.661, 0.709] 0.488 [0.456, 0.519] 0.758 [0.716, 0.803] 0.618 [0.556, 0.676]
CONTES 0.715 [0.694, 0.736] 0.500 [0.467, 0.529] 0.799 [0.760, 0.835] 0.616 [0.553, 0.675]
HONOR 0.737 [0.716, 0.759] 0.531 [0.499, 0.563] 0.778 [0.738, 0.814] 0.578 [0.519, 0.640]
C-Norm 0.777 [0.755, 0.797] 0.604 [0.574, 0.635] 0.881 [0.855, 0.907] 0.700 [0.643, 0.755]
  1. The highest scores are in italics