Stock Market Trend Prediction Model Using Deep Learning Based Sentiment Analysis of Financial Data

Creating a forecasting model to identify the trend in the stock market is difficult due to the complex nature and behavior of investors participating in the stock market. The existing models and studies have mainly used past prices and other technical indicators based on past prices or volume to predict future prices or trends instead of Behavioral finance for the Indian market. This study demonstrates a novel neural network-based stock market trend prediction model, which employs sentiment analysis data of published financial information. Deep learning-based sentiment analysis applications are used to identify investor sentiment based on published financial information on Macroeconomic and Microeconomic factors. The deep learning-based trend prediction model using sentiment analysis data has a more than 96% accuracy rate. The study also provides further evidence of behavioral finance theory of herd mentality in the Indian market. The result of research will benefit in the field of financial investment and risk management in the Indian stock market.