To assist enterprises in making personalized financial decisions, this paper designs a big data-based financial decision support platform based on the design concept of “data processing-data analysis-data presentation-data decision-making,” providing a decision support environment for financial decision-makers. To optimize personalized financial decisions, a random forest algorithm is used to construct an enterprise financial data risk warning model. Sample data and financial risk warning indicators are selected, and the random forest algorithm is used to estimate feature importance. The confusion matrix is employed as the metric standard for financial warning results. The hyperparameters of the random forest model are optimized, including n-tree optimization and mtry selection. Financial indicator data from T-1 year, T-2 year, and T-3 year are extracted separately for risk prediction analysis, and corresponding random forest classification models are constructed based on this. Compare the financial risk prediction accuracy rates of each model to validate the feasibility of the random forest algorithm as a key technology for a big data financial decision support platform. For T-1 year data, the enterprise financial data risk warning model based on the random forest algorithm demonstrates the best predictive performance, with accuracy and recall rates both exceeding 90%, with accuracy as high as 97%.
