Optimization of Financial Budget and Design of Budget System for Vocational Colleges Based on Data Analysis

The traditional financial budget management in vocational colleges is generally featured by inadequate data integration, weak analytical abilities, incorrect budget planning, and ineffective management of budget execution monitoring and dynamic adjustment. In order to solve the above problems, a data-driven financial budget management and optimization system in vocational colleges is designed in this study. The system takes into account budget planning, budget execution, monitoring, adjustment, and performance analysis into one management process. Multi-source financial and business data are extracted, cleaned, transformed and loaded into the financial data warehouse through the ETL process. The multidimensional and tabular analytical models are built for analyzing financial overview, financial income and expenditure, balance analysis, comparison analysis, financial risk warning and business management. Moreover, an ensemble decision tree regression model with randomized feature selection and split point generation is proposed to predict major financial indices. In particular, quarterly final-account data and annual budget data of a vocational college are used to evaluate the proposed model. The experimental results show that the prediction accuracies of the six major financial indices including operating revenue, operating cost, operating profit, net cash flow, accounts receivable and accounts payable vary from 85.8% to 97.5%. Meanwhile, the corresponding cross validation scores range from 0.90 to 0.96. As can be seen from the learning curve results, the model has relatively stable generalization ability with the increase of training sample sizes. Therefore, the proposed system can help realize integrated financial data management, multidimensional visualization analysis, financial budget forecast and dynamic adjustment of budgets.