Table 5.

Cross validation of the rule sets. The fungus testing dataset was used to evaluate the rule sets generated from the A, B, C, D, and E training datasets.

	A rule set		B rule set		C rule set		D rule set		E rule set
	precision	confidence	precision	confidence	precision	confidence	precision	confidence	precision	confidence
A testing data	71%	69%	72%	69%	59%	56%	46%	42%	52%	48%
B testing data	68%	67%	76%	74%	61%	59%	43%	41%	51%	49%
C testing data	72%	68%	69%	65%	87%	85%	79%	76%	66%	62%
D testing data	66%	61%	43%	38%	74%	72%	88%	85%	68%	65%
E testing data	52%	50%	65%	62%	60%	58%	64%	62%	77%	75%

A: actinobacteria B: bacillales C: fungi D: nematode + arthropoda E: viridiplantae