Application of Multilabel Learning Using the Relevant Feature for Each Label in Chronic Gastritis Syndrome Diagnosis

Table 3

Performance of different multilabel learning algorithms.

Group (mean±std)	ML-kNN	ECC	BSVM	BP-MLL	RANK-SVM	REAL
Average precision	0.759±0.029	0.802±0.016	0.802±0.016	0.540±0.023	0.707±0.022	0.820±0.029
Coverage	0.200±0.023	0.186±0.019	0.174±0.023	0.345±0.039	0.237±0.016	0.160±0.020
Hamming loss	0.167±0.014	0.148±0.016	0.156±0.014	0.304±0.014	0.214±0.014	0.142±0.019
One error	0.375±0.050	0.261±0.024	0.307±0.022	0.755±0.029	0.449±0.034	0.283±0.055
Ranking loss	0.167±0.025	0.190±0.025	0.130±0.017	0.334±0.040	0.206±0.014	0.117±0.018