This is the authors' abstract. We don't add key points for this paper.
Stock trend forecasting is crucial to maximizing returns from stock investments. However, the stock market is affected by various factors, resulting in uncertainty and volatility in stock prices. People perform trend prediction analysis by extracting emotional features from textual information and integrating them with price data to enhance the precision of capturing stock market trends. Prior research has encountered several challenges, including inadequate emotional feature extraction and limitations in integrating diverse structured data. Therefore, this paper constructs a dual-channel input stock trend prediction model to enhance prediction accuracy. Initially, the model utilizes the BERT to convert review text into a vectorized representation. Subsequently, a TCN is employed to further extract the emotional features in the text. Additionally, a CNN is leveraged to extract stock price features. Finally, the model integrates the textual emotional features with price features. Experimental conducted on PingAn Bank and Wuliangye stock data revealed prediction accuracies of 72.26% and 76.77% respectively. Compared with SVM-LSTM and EMD-emotion-LSTM, the proposed model demonstrated varying degrees of improved accuracy.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Chen et al. (2024) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: