Table 4.
Classification accuracies of shallow machine learning algorithms to predict whether the honey-water blending mixture was mixed or non-mixed. E—Energy, SAA—Sum Absolute Amplitude, G—Gradients of Features, PCs—Principle Components, DWT—Discrete Wavelet Transform.
| ANN (% Correct) | SVM (% Correct) | LSTM (% Correct) | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Features | Non-central | Central | Combined | Non-central | Central | Combined | Non-central | Central | Combined |
| E, SAA | 65.5 | 85.0 | 83.4 | 80.2 | 80.1 | 86.0 | 89.5 | 93.0 | 88.0 |
| E, SAA, G | 76.1 | 91.1 | 92.5 | 77.2 | 91.1 | 89.1 | 89.8 | 96.3 | 95.4 |
| PCs | 76.2 | 90.7 | 83.1 | 75.1 | 81.8 | 80.4 | 82.9 | 86.1 | 91.5 |
| PCs, G | 79.3 | 93.0 | 92.7 | 71.8 | 83.7 | 86.7 | 86.2 | 89.4 | 93.7 |
| DWT, E, SAA | 79.2 | 91.7 | 90.6 | 72.1 | 82.3 | 86.4 | 77.8 | 95.1 | 92.9 |
| DWT, E, SAA, G | 80.9 | 92.4 | 94.6 | 82.9 | 92.1 | 91.9 | 80.8 | 94.5 | 90.8 |
| DWT, PCs | 71.9 | 88.7 | 90.0 | 76.2 | 82.7 | 80.1 | 82.5 | 84.5 | 89.5 |
| DWT, PCs, G | 80.5 | 95.0 | 93.9 | 75.0 | 85.3 | 91.0 | 79.7 | 86.1 | 90.4 |