Research on System Strategy Optimization Based on Random Forest Algorithm

The Random Forest model, a supervised learning algorithm comprising numerous decision trees, is known for its high accuracy and generalization capabilities. This paper extends its application by employing the Random Forest algorithm in a strategy-based scenario addressing system sustainability under extreme conditions. The goal is to establish a sustainable evaluation framework using a fuzzy integrated assessment approach based on Random Forest. First, a prediction model was developed to analyze system performance and identify the influence of various factors through the Random Forest algorithm. Subsequently, a classification process was implemented using the AHP index model to categorize key structural components into different groups such as cultural, historical, economic, and functional entities. Finally, the fuzzy comprehensive evaluation method was applied to determine the sensitivity of different structural types to various stress indicators under extreme conditions. This study provides novel computational perspectives and methodological tools for optimizing strategies in complex system scenarios, offering a reference for addressing similar optimization challenges.