Title: Prediction of economic benefits of market digital transformation based on federal learning algorithm
Authors: Lu Zhang; Wanqing Chen; Hengzhi Nie
Addresses: Department of Economics and Trade, Henan Polytechnic Institute, Nanyang, 473001, Hennan, China; Department of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, 430073, China ' Zhengzhou Fengyang Foreign Language School, Zhengzhou, 450000, Hennan, China ' The Academy of Digital China (FuJian), Fuzhou, 350003, China
Abstract: In order to solve the problems of low prediction accuracy and long prediction time existing in traditional methods for predicting the economic benefits of market digital transformation, a method for predicting the economic benefits of market digital transformation based on federal learning algorithm is proposed. Obtain the prediction indicators and build a prediction indicator system for the economic benefits of market digital transformation. Pre-process the indicator data using the principal component analysis method, build a sample dataset, build a federal optimisation algorithm using the random gradient descent method, establish a minimum loss function, build a prediction model for the economic benefits of market digital transformation, and output the prediction results. The experimental results show that the prediction accuracy of the proposed method can reach more than 95%, and the prediction time is always kept within 3 s, with good prediction effect and efficiency.
Keywords: federated learning algorithm; digital transformation of the market; economic performance; principal component analysis; random gradient descent method.
International Journal of Sustainable Development, 2023 Vol.26 No.3/4, pp.190 - 205
Received: 30 Dec 2022
Accepted: 30 Apr 2023
Published online: 20 Oct 2023 *