Table 6.

Performance analysis of model-based reasoning methods in combination with rule-based reasoning methods for each syndrome element in the test data set with corpus 1 and corpus 2.^a

Syndrome element	Corpus 1					Corpus 2
	Precision	Recall	F1 score	Support	Precision		Recall	F1 score	Support
Phlegm	0.9907	0.9538	0.9719	1233	0.9935		0.9951	0.9943	1233
Wind	0.9926	0.9218	0.9559	435	0.9953		0.9770	0.9861	435
Cold	0.9800	0.9722	0.976	503	0.996		1.000	0.998	503
Heat	0.9704	0.8903	0.9286	811	0.9675		0.9174	0.9418	811
Qi-deficiency	0.9616	0.9756	0.9686	616	0.9871		0.9935	0.9903	616
Yin-deficiency	1.000	0.9851	0.9925	403	0.9975		0.9801	0.9887	403
Lung	1.000	1.000	1.000	2815	1.000		1.000	1.000	2815
Spleen	0.9644	0.9457	0.955	258	0.9771		0.9922	0.9846	258
Kidney	0.9882	0.9825	0.9853	171	0.9826		0.9883	0.9854	171
Average (weighted)	0.9885	0.968	0.9779	7245	0.9922		0.9863	0.9892	7245

^aCorpus 1 consists of syndrome and sign information, and corpus 2 consists of syndrome and sign information plus clinical diagnosis information. The average accuracy was 0.9229 (95% CI 0.9099-0.9319) for syndrome pattern in the test data set with corpus 1, and 0.9559 (95% CI 0.9429-0.9699) for syndrome pattern in the test data set with corpus 2.