Through a review of the C4.5 Decision Tree, Random Forest, Support Vector Machine and BP Neural Network algorithm and the obtained in the study of TCM syndrome differentiation, a compound structure of intelligent syndrome differentiation and formula selection model was designed. The model was implemented and tested. Results showed that the accuracy of the corresponding results was higher than that of the single algorithm, laying the foundation for the model which was suitable for the complex Pathogenesis of clinical diagnosis and treatment of auxiliary system.