Table 6 Statistical significance of differences in classification accuracy between binary combination of classfication algorithms and proposed Hybrid algorithm
From: A novel hybrid model to predict concomitant diseases for Hashimoto’s thyroiditis
Classification methods | Proposed hybrid method | Z-test | \(x^2\)-test | p value |
|---|---|---|---|---|
SVM-RF | Hybrid KNN-MLP | 7.02 | 49.28 | \(<0\).0001 |
SVM-LR | Hybrid KNN-MLP | 5.72 | 32.71 | \(<0\).0001 |
SVM-KNN | Hybrid KNN-MLP | 12.75 | 162.56 | \(<0\).0001 |
SVM-MLP | Hybrid KNN-MLP | 8.98 | 80.64 | \(<0\).0001 |
RF-LR | Hybrid KNN-MLP | 6.72 | 45.16 | \(<0\).0001 |
RF-KNN | Hybrid KNN-MLP | 3.24 | 10.50 | \(=0\).0007 |
RF-MLP | Hybrid KNN-MLP | 8.1 | 65.61 | \(<0\).0001 |
LR-KNN | Hybrid KNN-MLP | 7.65 | 58.52 | \(<0\).0001 |
LR-MLP | Hybrid KNN-MLP | 8.72 | 76.03 | \(<0\).0001 |
KNN-MLP | Hybrid KNN-MLP | 4.01 | 16.1 | \(<0\).0001 |