Author(s)

Xinyu Liu, Junyan Liu, Yi Zhang, Yuxin Sun, Zhiyang Wang, Yuqi Wu, Haodong Hong, Depei Zhang

Corresponding Author

Xinyu Liu

ABSTRACT

Fiscal revenue is not only an important source of national revenue, but also the basis for reflecting the country’s economic conditions. The trend of fiscal revenue determines the development trend of the local economy and market prospects, and is particularly important for the development of my country’s financial industry. For this reason, this article applies support vector machines to the forecast of fiscal revenue. In order to solve the shortcomings of time-consuming, labor-intensive and inefficient selection of model parameters of support vector machines, this paper proposes a method to optimize the selection of the penalty parameters and kernel parameters of the support vector machine using the Fruit fly optimization algorithm of global optimization, and establishes a prediction model based on the Fruit fly optimization algorithm to optimize the support vector machine. The simulation results show that the model optimized by Fruit fly optimization algorithm has higher prediction accuracy and meets the demand for prediction accuracy.

KEYWORDS

Fruit fly optimization algorithm, Support vector machine, fiscal revenue, forecasting model,