| 摘要 |
The apparatus for selecting an optimal decision tree for data mining, comprising input means through which multidimensional data Dn and a plurality of nested decision trees Ti created by using the multidimensional data are inputted calculating means for calculating a probability PrDn‖Ti of each of the multidimensional data for a given tree by using the multidimensional data and the plurality of nested decision trees inputted through the input means a prior probability calculating means for calculating a prior probability PrTi for the given tree posterior probability calculating means for calculating a posterior probability PrTi‖Dn based on Bayesian theorem by using the probability PrDn‖Ti of each of the multidimensional data for the given tree and the prior probability PrTi for the given tree and decision tree selecting means for selecting a decision tree with a largest posterior probability as a single optimal decision tree. |
