Prediction of non-suicidal self-injury (NSSI) among rural Chinese junior high school students: a machine learning approach

Table 3 Metrics for each non-suicidal self-injury (NSSI) predictive model in the test set

Value	Support vector machine				Decision tree	Random forest
Value	Linear	Radial	Polynomial	Sigmoid	Decision tree	Random forest
Sensitivity	0.32	0.26	0.34	0.54	0.59	0.65
Specificity	0.33	0.45	0.44	0.23	0.76	0.72
AUC	0.76	0.73	0.67	0.64	0.69	0.76
Accuracy	0.33	0.38	0.40	0.34	0.70	0.70
Precision	0.22	0.22	0.26	0.29	0.58	0.57
F1 Score	0.26	0.24	0.29	0.37	0.59	0.61

Legend The contents of this table are metrics for model testing (using the test set)

ISSN: 1744-859X