Computational Prediction of Subjective Human Immunodeficiency Virus Status in Malawi Using a Random Forest Approach

Table 2

Prediction results of the random forest method (confusion matrix) in the present study.

Data	Categories	Sensitivity (%)	Specificity (%)	Accuracy (%)	MCC (%)
Training	Both Sex	86.59	99.94	97.24	76.47
Testing		93.79	99.97	98.63	74.83
Training	Males	88.96	100.00	97.81	78.06
Testing		95.72	99.71	98.92	69.39
Training	Females	89.78	99.87	97.73	75.11
Testing		94.70	100.00	98.96	73.02

Source. MLSFH, 1998–2010. Note. All outputs were significant at .