Title: Evaluation on stock market forecasting framework for AI and embedded real-time system

Authors: Yu Lin

Addresses: School of Statistics, Southwestern University of Finance and Economics, Chengdu 610074, Sichuan, China; Joint Lab of Data Science and Business Intelligence, Southwestern University of Finance and Economics, Chengdu 610074, Sichuan, China

Abstract: Since its birth, the stock market has received widespread attention from many scholars and investors. However, there are many factors that affect stock prices, including the company's own internal factors and the impact of external policies. The extent and manner of fundamental impacts also vary, making stock price predictions very difficult. Based on this, this article first introduces the research significance of the stock market prediction framework, and then conducts academic research and analysis on two key sentences of stock market prediction and artificial intelligence in stock market prediction. Then this article proposes a constructive algorithm theory, and finally conducts a simulation comparison experiment and summarises and discusses the experiment. Research results show that the neural network prediction method is more effective in stock market prediction; the minimum training rate is generally 0.9; the agency's expected dilution rate and the published stock market dilution rate are both around 6%.

Keywords: stock market forecast; embedded real-time system; artificial intelligence; back propagation neural network; dilution rate.

DOI: 10.1504/IJDMB.2024.139448

International Journal of Data Mining and Bioinformatics, 2024 Vol.28 No.3/4, pp.219 - 235

Received: 28 Feb 2023
Accepted: 08 Sep 2023
Published online: 02 Jul 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article