In the stock market, stock prices are often influenced by macroeconomic indicators, market sentiment, company fundamentals, and other factors, presenting complex nonlinear relationships. However, traditional stock trend prediction methods are often affected by factors such as data noise and nonlinear relationships, resulting in low prediction accuracy. Therefore, a research on financial market stock trend prediction method based on support vector machine is proposed. Firstly, web scraping technology is used to obtain financial market stock data, including historical price data, trading volume data, and technical indicator data, and these data are preprocessed. Then, linear discriminant analysis is used to extract data features, and a support vector machine prediction model is constructed. The extracted features are used as input data to obtain predicted stock trends in the financial market. The experimental results show that the stock prediction trend of the proposed method is consistent with the actual trend, and the mean square error is small, which has practical application value.

Liu Linfeng, Dong Mei. Research on Stock Trend Prediction Method in Financial Markets Based on Support Vector Machines. International Journal of Frontiers in Sociology (2024), Vol. 6, Issue 6: 126-131. https://doi.org/10.25236/IJFS.2024.060620.

[1] Yang Longguang. Prediction of intraday stock price trends based on micro market structure [J]. Journal of Southwest Minzu University (Humanities and Social Sciences Edition), 2023,44 (10): 122-130

[2] Sun Ziyu, Ren Ran, Wei Xizhe. Research on Stock Classification and Prediction Based on DTW-TCN [J]. Computer and Modernization, 2023 (08): 31-37

[3] Cui Ting, Huang Feiran. Stock prediction based on sentiment analysis model: a prediction model combining GRU and ALBERT [J]. Dongyue Tribune, 2024, 45 (02): 113-123