Abnormal Stock Price Fluctuation Recognition Based on Convolutional Neural Networks and Its Application in Financial Risk Monitoring

Abstract

This paper studies the abnormal stock price fluctuation recognition algorithm based on convolutional neural network (CNN). Through comparative experiments with models such as support vector machine (SVM), long short-term memory network (LSTM), and bidirectional long short-term memory network (BiLSTM), the superior performance of CNN in the task of identifying abnormal stock price fluctuations is verified. The experimental results show that CNN is superior to other models in terms of accuracy and F1 score, and can more effectively capture the local fluctuation characteristics of stock prices. This study provides technical support for abnormal fluctuation monitoring and risk warning in the financial market, and lays the foundation for further application of deep learning in financial analysis.