Table 4 The total accuracy and Kappa index obtained from three Neural Network models on 13 datasets (FCdb and 12 datasets that obtained from attribute weighting models)
From: Prediction of lung tumor types based on protein attributes by machine learning algorithms
Data Base | Auto MLp Accuracy | Neural Net Accuracy | Perceptron Accuracy | Auto MLp Kappa | Neural Net Kappa | Perceptron Kappa |
|---|---|---|---|---|---|---|
Chi Squared | 73.79% | 70.23% | 54.09% | 56.39% | 51.54% | 24.84% |
Info Gain Ratio | 80.76% | 83.41% | 52.58% | 68.53% | 71.55% | 13.60% |
FCdb | 69.24% | 81.59% | 50.76% | 50.05% | 69.36% | 3.43% |
SVM | 85.15% | 87.73% | 57.80% | 75.66% | 79.66% | 20.61% |
Uncertainty | 82.58% | 81.59% | 52.42% | 71.33% | 69.92% | 18.58% |
PCA | 51.67% | 51.67% | 30.98% | 0.77% | 0.00% | -5.04% |
Relief | 77.27% | 75.61% | 51.67% | 62.52% | 60.30% | 16.09% |
Rule | 76.06% | 80.53% | 48.03% | 60.96% | 67.58% | 5.45% |
Deviation | 51.67% | 52.50% | 30.98% | 1.73% | 3.33% | -5.04% |
Gini Index | 76.29% | 76.21% | 48.86% | 61.62% | 61.97% | 11.38% |
Info Gain | 85.91% | 85.98% | 51.44% | 76.98% | 77.09% | 16.98% |
SAM | 66.32% | 64.62% | 52.65% | 45.56% | 42.92% | 15.51% |
MR | 76.29% | 75.45% | 58.56% | 61.88% | 60.12% | 25.92% |