Research on the Application of Decision Tree Algorithm in the Selection of Variables in Cost Estimation Model

Abstract

In order to solve the problem that there may be too many input variables, which affect the convergence speed of the model and the easy formation of over fitting, this paper proposes to use the decision tree algorithm to search for the best grouping variables when constructing the decision tree. Explore a new method to filter model input variables. Through the case analysis, the precision of the forecast model is satisfied by using the decision tree algorithm. This method will be an important knowledge method reserve in the processing of specific engineering problems. Especially for complex problems such as low-dimensional analysis, rapid modeling will play an important role.