Diabetes Detection Models in Mexican Patients by Combining Machine Learning Algorithms and Feature Selection Techniques for Clinical and Paraclinical Attributes: A Comparative Evaluation

Table 9

Performance of the classification models implemented with the 14-feature subset obtained using the Akaike criterion as a feature selection technique.

Metric	SVM	RF	kNN	GB	ET	NB
AUC	0.98	0.98	0.95	0.98	0.98	0.97
Specificity	0.93	0.94	0.85	0.95	0.94	0.91
Sensitivity	0.97	0.98	0.95	0.95	0.97	0.95
Accuracy	0.95	0.96	0.89	0.95	0.95	0.93
score	0.95	0.95	0.89	0.95	0.95	0.93
Precision	0.97	0.98	0.94	0.96	0.97	0.95