Analysis and Prediction of Stock Prices of Commercial Companies Using Orange Data Mining Software

Stock market prediction remains a challenging task due to the high volatility and complexity of financial data. This study investigates the application of machine learning models for analyzing and forecasting stock prices, leveraging the Orange Data Mining platform. Unlike traditional forecasting methods, such as ARIMA and technical indicators (RSI, MACD), machine learning techniques capture nonlinear dependencies in financial data, leading to improved predictive accuracy. The research implements multiple machine learning models, including Support Vector Machines (SVM), k-Nearest Neighbors (kNN), Decision Trees, Random Forest, and Artificial Neural Networks (ANNs). The models are trained and evaluated using historical stock data, with Tesla, Inc. serving as a case study. Performance is assessed through statistical metrics such as Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and R2. Additionally, a comparative analysis highlights the advantages of machine learning approaches over conventional methods in terms of predictive accuracy and adaptability to market fluctuations. The results demonstrate that machine learning-based approaches, particularly Decision Trees and Neural Networks, outperform traditional statistical models in forecasting stock prices. The study underscores the potential of integrating visual-based data mining platforms like Orange for financial market analysis, offering a user-friendly and effective approach for traders and analysts. The proposed methodology can be extended to other financial assets and industries, emphasizing the role of artificial intelligence in modern stock market forecasting.