Table 12 Comparison with deep learning models on the EHRs diabetes prediction dataset.
From: A deep learning framework with hybrid stacked sparse autoencoder for type 2 diabetes prediction
Models | Pre | Rec | F1 | Acc | AUC | Hamming loss |
|---|---|---|---|---|---|---|
CNN | 0.79 | 0.54 | 0.65 | 0.70 | 0.77 | 0.30 |
LSTM | 0.82 | 0.65 | 0.72 | 0.75 | 0.85 | 0.25 |
SSAE | 0.83 | 0.70 | 0.75 | 0.77 | 0.87 | 0.23 |
HSSAE (Proposed) | 0.92 | 0.95 | 0.94 | 0.93 | 0.99 | 0.07 |
