Ensemble of Rotation Trees for Imbalanced Medical Datasets
Pseudocode 1
Pseudocode of ensemble of rotation trees for medical datasets.
Training:
Input:
Xa—the abnormal set,
Xn—the normal set,
M—the number of classifiers in the ensemble
Output:
the ensemble H with M classifiers
Begin:
1. ;
2. H =∅;
3. repeat
4. sample a subset Dn from ;
5. ; //balanced dataset
6. Split F into subsets: ,;
7. ;
8. repeat
9. Let Di, j be the data set of Di for the feature in Fi, j;
10. Select a bootstrap sample subset D’i, j from Di, j of size 50% of the number of objects in Di, j。 Denote as the new set;
11. Apply PCA on Fi, j and D’i, j to obtain the coefficients in a matrix Ri, j;
12. until j = n/L
13. Arrange the Ri, j in a rotation matrix Ri as in equation (1), ; //refer to (1)
14. Di, train = DiRi; //obtain novel dataset through projecting balanced dataset Di to the new space defined by Ri
15. Build classifier hi using Di, train; //learn classifier on the novel balanced dataset Di
16. ;
17. until i = M
18. return H
Classification Phase:
For a given x, let hi(Rix) be the probability assigned by the classifier hi to the hypothesis that x comes from class ωj. Calculate the confidence for each class that x belongs to using the average combination method:
Assign x to the class with the largest confidence.