An Intelligent Stock Market Recommendation System Based on Financial Behaviour and Investor Profiling

Choosing suitable stocks can be difficult for individual investors, especially those without professional financial knowledge. Many stock-analysis approaches focus on market prices, technical indicators or company fundamentals, while an investor's financial behaviour, risk tolerance and investment horizon may receive less attention. This paper presents an intelligent stock market recommendation framework that connects investor profiling with machine-learning-based analysis. The project materials describe a tabular dataset containing 16,203 records and 36 columns. The workflow includes data inspection, exploratory data analysis, feature preparation, investor-profile classification and recommendation analysis. The project materials also report an Extra Trees result of 77.4% accuracy and a weighted F1-score of 0.748 using an 80:20 train-test split. These figures are treated as reported project results and must be reproduced from the final implementation before being considered verified. The dataset has an uneven distribution of investor profiles and recommendation labels, making per-class evaluation important. The framework is intended as decision support, not as a guarantee of future stock performance or investment returns.