Abstract
In order to ensure the benign operation of the social security fund system, it is necessary to understand the social security fund facing all aspects of the risk, more importantly to know the relationship between different risks. Based on RBF, the interpretative structure model is applied to draw the risk correlation hierarchy diagram, which provides a scientific risk management method for the social security fund. RBF neural network is used to build the risk warning model of social security fund operation. Then, put forward the corresponding risk treatment scheme to the warning signal. Finally, the RBF neural network is used for comprehensive risk warning. In this paper, the risk warning of social security fund operation is the research object, and the corresponding risk treatment scheme is put forward for the warning signal. This paper uses an improved ant colony algorithm to optimize the parameters of the RBF neural network, which overcomes the shortcomings of the traditional RBF neural network such as slow convergence, ease of falling into local extremes, and low accuracy, and improves the generalization ability of the RBF neural network. It has the characteristics of good output stability and fast convergence speed. On this basis, the prediction model based on the improved ANT colony-RBF neural network is established, and the MATLAB software calculation tool is used for accurate calculation, which makes the prediction results of coal mine safety risk more accurate and provides more reliable decision basis for decision makers. The results show that the network has small calculation error, fast convergence, and good generalization ability.
Highlights
China’s insurance industry is in an unprecedented transition period
Whether starting from artificial neural network technology or other technologies, want to go to, or technically is a breakthrough and innovation of traditional early warning system, it solves the traditional model is difficult to deal with high degree of nonlinear model, adaptive ability, lack of access to information and knowledge of indirect, time consuming and low ability of difficulties, so as to lay the foundations for the early warning go to practical [5,6,7,8]
The multilayer feedforward network is based on back propagation (BP), but most of the learning algorithms of this network must be based on some nonlinear optimization technology. erefore, its application is limited due to the large amount of computation and slow learning speed [11,12,13]
Summary
Received August 2021; Revised September 2021; Accepted 21 September 2021; Published 30 September 2021. RBF neural network is used to build the risk warning model of social security fund operation. Is paper uses an improved ant colony algorithm to optimize the parameters of the RBF neural network, which overcomes the shortcomings of the traditional RBF neural network such as slow convergence, ease of falling into local extremes, and low accuracy, and improves the generalization ability of the RBF neural network It has the characteristics of good output stability and fast convergence speed. Interpretive structural models can be used to describe and analyze the interrelationships between system elements It uses the known relationship between system elements to build the adjacencies matrix and generate the reachability matrix and generate the hierarchy diagram [1]. Is paper tries to analyze the risk structure of social security fund operation from the perspective of system engineering by using the interpretive structure model [2]. Erefore, it is necessary to establish a comprehensive risk early warning system for property insurance companies in China [3]. e research on the risk early warning system of property insurance company is relatively few, and it mainly draws lessons from the banking risk early warning method
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
