School of Economics and Finance, Xi'an Jiaotong University, Xian, Shaanxi 710061, China.
School of Business, Xi'an International University, Xian, Shaanxi 710077, China.
Comput Intell Neurosci. 2021 Nov 19;2021:3113584. doi: 10.1155/2021/3113584. eCollection 2021.
In the context of economic globalization and digitization, the current financial field is in an unprecedented complex situation. The methods and means to deal with this complexity are developing towards image intelligence. This paper takes financial prediction as the starting point, selects the artificial neural network in the intelligent algorithm and optimizes the algorithm, forecasts through the improved multilayer neural network, and compares it with the traditional neural network. Through comparison, it is found that the prediction success rate of the improved genetic multilayer neural network increases with the increase of the dimension of the input image data. This shows that, by adding more technical indicators as the input of the combined network, the prediction efficiency of the improved genetic multilayer neural network can be further improved and the advantage of computing speed can be maintained.
在经济全球化和数字化的背景下,当前金融领域正处于前所未有的复杂形势之中。应对这种复杂性的方法和手段正在朝着图像智能的方向发展。本文以金融预测为出发点,选取智能算法中的人工神经网络并对算法进行优化,通过改进后的多层神经网络进行预测,并与传统神经网络进行比较。通过对比发现,改进后的遗传多层神经网络的预测成功率随着输入图像数据维度的增加而增加。这表明,通过将更多技术指标作为组合网络的输入,可以进一步提高改进后的遗传多层神经网络的预测效率,并保持计算速度的优势。