A neural network-based quantification of risk in large health investments and its applicability to international investment law

The rapid development of the big health industry brings opportunities for investors, while the risks faced by the investment also increase day by day.This paper combines artificial neural networks and genetic algorithms to construct a risk evaluation model for big health investment based on GA-BP neural network, which provides a tool for analyzing and evaluating the risk of big health investment.The normalization function is used to standardize the model sample data, determine the learning rate, target error and other model parameters, and on the basis of the BP neural network, the genetic algorithm is used to optimize the connection weights and thresholds to achieve the quantitative function of the evaluation of the risk of big health investment.In order to verify the performance of the model to carry out large health investment risk quantification experiments, this paper's model convergence performance is better, the minimum Loss loss value is 0.37, the overall fluctuation is smaller.Compared with BP and PSO evaluation models, the risk evaluation accuracy of this paper's model is above 93.5%, and the comprehensive evaluation accuracy is higher than 91.1%, which can realize more accurate risk prediction, and its applicability in the international investment law has a high practical value.