Research Article
Research on a Machine Learning-Based Method for Assessing the Safety State of Historic Buildings
Table 3
Recognition accuracy of KCCA-SVM with different dimensions.
| Datasets | Dimension | Accuracy | Minimum value | Average value | Maximum value |
| ORL | 10 | 0.728 | 0.735 | 0.751 | 15 | 0.841 | 0.857 | 0.866 | 20 | 0.921 | 0.922 | 0.923 | 25 | 0.922 | 0.922 | 0.923 |
| Yale | 10 | 0.769 | 0.782 | 0.791 | 15 | 0.838 | 0.849 | 0.858 | 20 | 0.902 | 0.903 | 0.906 | 25 | 0.906 | 0.906 | 0.906 |
| AR | 10 | 0.842 | 0.855 | 0.861 | 15 | 0.882 | 0.897 | 0.908 | 20 | 0.943 | 0.952 | 0.958 | 25 | 0.950 | 0.956 | 0.958 |
| PIE | 10 | 0.812 | 0.821 | 0.833 | 15 | 0.863 | 0.872 | 0.884 | 20 | 0.887 | 0.896 | 0.901 | 25 | 0.907 | 0.913 | 0.916 |
|
|