A Model of Integrating Bert and BiGRU+ Attention Dual-channel Mechanism for Investor Sentiment Analysis of Stock Price Forecast

Abstract

Investor sentiment and emotions have a strong impact on financial markets. In recent years there has been increasing interest in analyzing the sentiment of investors for stock price prediction using machine learning. Existing prediction models mostly depend on the analysis of trading data and company profit. few prediction theories have been built based on individual investors' sentiments. The fundamental reason is the difficulty to measure individual investors' sentiment.