Applied Mathematics and Nonlinear Sciences (Jan 2024)
Construction of Intelligent Financial Supervision Mechanism for Colleges and Universities Based on RPA Technology
Abstract
The RPA financial monitoring mechanism utilizes a predictive model to capture and store university data in preprocessing, and adopts a decision tree algorithm to construct a specific predictive model with good classification ability, and its logical structure is very similar to the decision-making ideas of human beings in the natural environment. CNN-LSTM financial risk assessment model is also utilized to carry out scientific risk assessment, the output of CNN is used as the input of LSTM, combining the advantages of the two to improve the learning ability of the financial fraud risk assessment model, and dynamic early warning can be accomplished beforehand for potential risk points. By analyzing the accuracy and coverage of the financial violation prediction model, 10.28% of violations were found with 96.7% accuracy. The accuracy is lower when the proportion of financial violations is higher. Risk assessment of fundraising activities and working capital of universities can be found that Z private universities have a relatively single source of income, tuition income is dominant, and in 2019, 2020 and 2022, the proportion of constructive expenditures of Z private universities reaches more than 20%, while the expenses of scientific research costs are less, however, the university’s comprehensive early warning status is relatively good. The financial risk is not significant and is in a safer range. In the university research fund violation supervision, the weight of researchers still have balance funds after two years after the completion of the project is 0.073, which indicates that the research project fund management supervision is not in place.
Keywords