Predictive Models for the Medical Diagnosis of Dengue: A Case Study in Paraguay

Table 3

The performance of the ANN and SVM classifiers on 30 test datasets.

Classifiers	Accuracy (%)		Sensitivity (%)		Specificity (%)
	Mean	Variation	Mean	Variation	Mean	Variation
ANN-MLP	96	2	96	4	97	3
ANN-RBF	55	11	58	10	52	24
SVM linear	64	8	56	18	71	10
SVM Gaussian	86	9	84	11	89	10
SVM polynomial	92	14	93	12	92	16

The ANN-MLP classifier achieved better results with an average of 96% accuracy, 96% sensitivity, and 97% specificity, with low variation in thirty different partitions of the dataset. Also, the SVM polynomial classifier obtained results above 90% for the three indicators with acceptable variations.