Stock Price Prediction Based on FinBERT-LSTM Model

This study constructs a FinBERT-LSTM model to improve the accuracy of stock price prediction in the Chinese financial market. The FinBERT model, a Chinese language-specific, finance-oriented pre-trained model, is utilized to extract sentiment from financial texts. This sentiment data is combined with historical stock price data and input into an LSTM network to predict future stock prices. The model is trained and tested on the real data set with Kweichow Moutai as an example. The experimental results show that the FinBERT-LSTM model achieves better prediction accuracy than the traditional machine learning and deep learning models. This study provides a more accurate stock price prediction model by combining sentiment analysis with quantitative data, offering valuable insights for investors and researchers in the Chinese financial market.