A Decision Tree Method Based on Support Vector Machineto Identify Egg Embryo

This paper mainly studies a decision tree method based on support vector machine to identify egg embryo. It analyses the main characteristics of the types of egg embryo and sets up a kind of multilayer decision tree classifier by using "solution space". And it figures out the correct rate of the decision tree classifier, which concentrates on the types of egg embryo. By introducing the support vector machine (SVM) algorithm based on the structure of the binary tree for multi-class classification, it identifies different kinds of egg embryo. Not only does this method make the decision capability of the optimal one in every level of the decision tree, but also assures the overall optimal performance of the whole decision tree, and effectively improves the correct recognition rate of the decision tree classifier about the types of egg embryo.