Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
6904433 | Applied Soft Computing | 2016 | 41 Pages |
Abstract
- Supervised clustering is employed to generate a number of clusters.
- Clusters from different classes are pairwise combined to get a number of training subsets.
- The base classifiers constructed in the training subsets are diverse and locally accurate.
- Weighted voting mechanism is used to ensemble the base classifiers.
- The proposed approach improves the accuracy of credit scoring.
Related Topics
Physical Sciences and Engineering
Computer Science
Computer Science Applications
Authors
Hongshan Xiao, Zhi Xiao, Yu Wang,