Decision tree method based on KPCA and its application
Guoji Zhang · Jisuanji gongcheng yu sheji · 2007
Principal component analysis (PCA) is a popular data reduction technique for building decision tree. The complexity can be reduced and the classification precision of decision tree is improved. But it has drawbacks when it is used to solve nonlinear problem. Aimed at the foregoing point,a method of decision tree based on kernel principal component analysis (KPCA) is proposed. The ex-perimental result shows that it is feasible and effective. In comparison with PCA decision,there is also superior performance in KPCA decision tree.