AI-Powered Investment Advisor: Enhancing Financial Decisions with NLP and Predictive Analytics

Abstract

In recent years, Vietnam’s financial market has been assessed as high volatility, with stock prices often experiencing sudden shifts due to macroeconomic factors, regulatory policies, and investor sentiment. This complexity poses significant challenges for novice investors, who frequently struggle with analyzing market data and making informed decisions. Their limited financial literacy and restricted access to quality information further expose them to substantial risks and potential losses. To address these challenges, this study proposes the development of an AI-powered smart investment application capable of stock price prediction and portfolio optimization. A standout feature of the application is its intelligent chatbot, designed to provide real-time market insights, resolve stock-related queries, and offer technical support, thereby enhancing user experience. By automating decision-making processes, the application reduces errors, maximizes returns, and promotes transparency and fairness in financial markets. The system employs machine learning models to identify market trends and deep learning algorithms to improve stock price prediction accuracy using real-time market data. Additionally, natural language processing (NLP) enables the chatbot to understand natural language inputs and deliver relevant investment insights. The practical significance of this application lies in its ability to empower investors to make well-informed decisions with greater confidence while simultaneously improving their financial literacy. Furthermore, it fosters financial inclusion by expanding investment opportunities to a broader audience, including those with limited financial expertise, contributing to the sustainable development of Vietnam’s financial market. Future work will focus on enhancing real-time data processing speed, extending predictions to diverse financial assets, and integrating macroeconomic risk factors to increase the system's accuracy and applicability in complex market conditions