Optimization of Financial Market Forecasting Model Based on Machine Learning Algorithm

With the rapid growth of economic and social fields from digitalization to intelligence, the traditional risk management system is not suitable for the new model of financial innovation and development. Based on the advantages of machine learning, it is of great significance to use artificial neural network (ANN) to analyze the characteristics of financial markets. In this paper, the deep neural network model in machine learning algorithm is used for financial risk forecast, and a forecast model based on ANN is constructed. The model collects the relevant data of financial market, designs the deep learning network structure according to the training samples, trains the sample set by using the ANN-based forecast model, outputs the optimal solution, and completes the financial risk forecast. Simulation results show that, after many iterations, the accuracy of this method is obviously better than that of traditional Support Vector Machines (SVM) algorithm, with the accuracy as high as 96.84%, 20.55% higher than that of SVM algorithm, and the error reduced by 32.84%. Therefore, the financial market forecasting model based on ANN is a reasonable and feasible evaluation model.