Table 3.

Evaluation of the classification model for eight querying algorithms and two datasets on a small training set (with 16, 32, and 64 training samples) based on average AUC score and the standard deviation.

Dataset	Size of Training Set	LC	LCB	LCB2	LCMC	LCBMC	LCB2MC	IDD	Random
ASSERTION Dataset	16	71.52% ± 2.55%	79.16%±3.85%	80.91% ± 1.31%	76.87% ±5.89%	81.92% ± 1.41%	78.55% ± 4.73%	69.10% ± 2.70%	75.65% ±5.83%
	32	78.85% ± 4.30%	81.46% ± 2.77%	82.04% ± 1.38%	80.11% ± 1.91%	81.87% ± 2.45%	80.61% ± 1.31%	78.77% ± 2.55%	79.00% ±4.31%
	64	84.16% ± 1.42%	82.33% ± 2.05%	84.16% ± 0.74%	81.45% ± 2.34%	85.42% ± 1.03%	81.07% ± 1.45%	80.88% ±3.18%	83.12% ± 2.25%
NOVA Dataset	16	73.98% ± 6.25%	83.99% ± 4.93%	84.42% ±3.66%	82.01% ± 4.00%	83.98% ±5.81%	81.30% ±4.61%	78.91% ± 4.22%	76.70% ± 7.06%
	32	70.88% ± 2.96%	80.82% ±5.35%	85.33% ± 1.63%	81.52% ± 6.97%	85.69% ±3.05%	85.25% ±3.15%	77.27% ± 2.50%	79.03% ± 6.96%
	64	69.38% ±5.05%	91.79% ± 0.54%	91.82% ± 0.58%	92.21% ±0.71%	91.03% ± 2.80%	90.16% ± 1.04%	71.77% ± 2.63%	83.57% ± 4.88%