Principal Component Analysis-based Approach for Multivariate Decision Tree Construction

Most decision tree construction algorithms check up only one attribute on each node. This class of decision tree, called univariate decision tree, ignores the connection effect among the attributes inside the certain information system, which is actually widely occur. Furthermore, the cost of pruning is usually large. Aiming at the foregoing two defects, principal component analysis-based approach for multivariate decision tree construction is proposed in this paper. And several principal components should be extracted to constructing decision tree. The experiment results demonstrate it is a simple decision tree construction algorithm with high efficiency.