Table 5.

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

Syndrome pattern	Corpus 1				Corpus 2
	Precision	Recall	F1 score	Support	Precision	Recall	F1 score	Support
Qi-deficiency of lung and spleen	0.9363	0.9514	0.9438	247	0.9957	0.9665	0.9809	239
Qi-deficiency of lung and kidney	0.9362	0.9999	0.9670	176	0.9781	0.9944	0.9861	179
Yin-deficiency of lung	0.9777	0.9733	0.9755	225	0.9902	0.9999	0.9951	203
Wind-cold attacking lung	0.9943	0.9943	0.9956	176	0.9878	0.9999	0.9939	162
Wind-heat attacking lung	0.9899	0.9120	0.9494	216	0.9150	0.9826	0.9476	230
Cold wheezing	0.9724	0.9832	0.9778	179	0.9750	0.9653	0.9701	202
Deficiency of qi and yin	0.9934	0.9804	0.9868	153	0.9932	0.9932	0.9945	147
Hot wheezing	0.9051	0.9931	0.947	144	0.9563	0.9808	0.9684	156
Phlegm-heat obstruction in lung	0.9389	0.9021	0.9201	613	0.9357	0.9125	0.9240	606
Phlegm obstruction in lung	0.9183	0.9344	0.9263	686	0.9461	0.9407	0.9434	691
Average (weighted)	0.9477	0.9471	0.9470	2815	0.9586	0.9584	0.9584	2815

^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 0.9471 (95% CI 0.9382-0.9549) for syndrome pattern in the test data set with corpus 1, and 0.9584 (95% CI 0.9510-0.9655) for syndrome pattern in the test data set with corpus 2.