부산대학교 도서관

Recent years have seen significant studies into applying machine learning techniques for stock price prediction. However, most existing work in this field focuses on examining historical stock price data for forecasting stock prices, ignoring the role of public sentiments and mood on the stock market. In this work, a new approach is introduced that considers the sentiment factor along with the historical stock price to increase the precision of stock price forecasting. The proposed system includes a deep neural network that takes historical stock price information and sentiment analysis of news headlines as input characteristics. The system evaluates the framework's performance using the dataset of news headlines and stock prices of six different industries. There currently needs to be more literature in the field of stock price prediction, which incorporates both historical stock price data and sentimental data. Hence, the proposed hybrid approach for stock price prediction using sentiment and technical analysis significantly contributes to estimating the stock price.