Skip to main content

Table 9 Performance evaluation for DL model

From: An effective correlation-based data modeling framework for automatic diabetes prediction using machine and deep learning techniques

Performance parameter

Deep CNN model

Evaluation without data modeling

Evaluation with data modeling

Accuracy (%)

88.38

96.13

Precision (%)

83.33

94.44

Recall (%)

83.34

94.42

F-Measure (%)

83.33

94.46

Specificity (%)

91.09

97.03

Sensitivity (%)

83.02

94.45

ROC area

0.872

0.959

False Positive Rate (FPR)

0.089

0.029

False Negative Rate (FNR)

0.166

0.055

Matthews Correlation

Coefficient (MCC)

0.744

0.914