Combining AHP and Decision Tree to Optimize Network Compliance Detection in Low-carbon Economy Financing Risk Assessment

Abstract

Low-carbon project financing risks are complex and volatile, affected by multiple factors such as policy, market, technology and a single assessment method, making it difficult to comprehensively capture compliance risks. In this paper, the Analytic Hierarchy Process (AHP) is combined with decision tree analysis to optimize network compliance detection in low-carbon economy financing risk assessment. AHP is first used to cascade the risk factors, determine the risk factor weights and perform sensitivity analysis. A decision tree model is then constructed using the CART (Classification and Regression Trees) algorithm, where identified risk factors are used as nodes and subdivided into nodes to form a hierarchical structure for systematic analysis. Finally, a real-time data update mechanism is established to integrate data from multiple sources, and the frequency of updates is set according to the environment. The model adjusts the decision path based on the new data and sets dynamic thresholds to respond to changes in risk after optimization using multiple decision trees. The study demonstrates that the identification accuracy of financing risk detection after model optimization reaches 85.89%, and the efficiency of compliance detection is improved by 47.06%. The combination of AHP and decision tree can effectively identify low-carbon economy financing risks and adapt to the fluctuating environment, meeting the financing needs of low-carbon economy.