Application of Decision Tree Random Forest Algorithm in the Evolution of Dual Innovation in SMEs

This study proposes an innovation path prediction and decision model based on decision tree and random forest algorithm for the evolution process of dual innovation in SMEs. First, the decision tree algorithm is used to analyze the key decision points of enterprises in the dual innovation process, revealing the key factors affecting innovation decisions and their interrelationships. Then, combined with multi-dimensional enterprise data, a multi-level and complex prediction model is constructed using the random forest algorithm to simulate the innovation evolution path of enterprises in different situations. Through simulation analysis of actual data of multiple SMEs, the model successfully predicts the choice tendency of enterprises in the dual innovation path. During the simulation process, the prediction accuracy of the decision tree model was 78.3 %, while the prediction accuracy of the random forest model increased to <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{9 2. 5 \%}$</tex>. This result shows that the random forest algorithm can more accurately capture the nonlinear relationship in the data and significantly improve the prediction accuracy of the model. Especially when faced with complex innovation path selection and multi-dimensional data, the random forest algorithm shows obvious advantages in accuracy, stability and generalization ability compared with the traditional decision tree algorithm.