Research Article
A Novel Brain Network Construction Method for Exploring Age-Related Functional Reorganization
Table 5
SVM accuracy in the optimized feature set.
| Density | 8% | 9% | 10% | 11% | 12% | 13% | 14% | 15% | 16% |
| | | | | | | | | | | 90 | 0.786 | 0.893 | 0.821 | 0.821 | 0.786 | 0.714 | 0.786 | 0.786 | 0.750 | 95 | 0.857 | 0.821 | 0.821 | 0.821 | 0.750 | 0.750 | 0.750 | 0.679 | 0.679 | 100 | 0.750 | 0.821 | 0.857 | 0.643 | 0.750 | 0.714 | 0.679 | 0.679 | 0.679 | Classic | 0.714 | 0.643 | 0.679 | 0.714 | 0.714 | 0.643 | 0.643 | 0.643 | 0.679 |
|
|
Accuracy was significantly higher in the SNV method than in the classic method (permutation test, at ; at ).
|