Enhancing Financial Market Predictions: Integrating Sentiment Analysis and Historical Data within ChatGPT
Project Author
Issue Date
2024-03-22
Authors
Turner, Jack
Loading...
Embargo
First Reader
Jumadinova, Janyl A.
Additional Readers
Kapfhammer, Gregory
Keywords
item.page.distribution
Abstract
This thesis explores the enhancement of financial market analysis through the integration of sentiment analysis and historical stock data predictions within the ChatGPT framework. Addressing the complexity and dynamic nature of financial markets, the research aims to develop a more accurate and holistic approach to stock market forecasting by combining quantitative historical data and qualitative sentiment insights derived from financial news. Utilizing Python libraries such as pandas, NumPy, yfinance, and natural language processing techniques, the project constructs a predictive model that analyzes historical stock trends alongside market sentiment. The innovation of this research lies in its integration within ChatGPT, facilitating an interactive tool that provides personalized financial insights. The experiments demonstrate the model’s potential in offering improved predictive accuracy and valuable market insights, suggesting a step towards the continued integration of AI in financial analysis. The research contributes to the field by providing a new approach to financial market analysis, merging quantitative historical data with qualitative sentiment insights within ChatGPT, thus offering a comprehensive framework for understanding market dynamics.
Description
Chair
Major
Computer Science
Department
Computer and Information Science