AI Stock Analytics and Forecasting Tool

AI Stock Price Prediction Dashboard Overview This project is an AI-powered stock analysis and prediction system developed as part of an MSc Data Science and Big Data Analytics final year project. The application uses multiple machine learning and time-series models to analyze historical stock data and predict future price movements. The system provides an interactive Streamlit dashboard where users can select a stock and choose different prediction models to generate a BUY / SELL / HOLD signal along with predicted price values. The dashboard also visualizes historical stock performance with technical indicators and displays predictions in Indian Rupees (₹). Features Interactive Streamlit dashboard Stock data retrieval using Yahoo Finance Visualization of historical stock performance Moving average indicators (MA50 & MA200) Multiple prediction models: Random Forest ARIMA LSTM Hybrid Model (combination of models) BUY / SELL / HOLD signal generation Currency conversion from USD to INR Model selection and stock selection directly from the dashboard Project Architecture User Input (Streamlit Dashboard) │ ▼ Stock Data Collection (yfinance API) │ ▼ Feature Engineering │ ▼ Prediction Models ├── Random Forest ├── ARIMA ├── LSTM └── Hybrid Model │ ▼ Prediction Output │ ▼ Dashboard Visualization Project Structure stock_ai_project │ ├── models │ ├── random_forest_model.pkl │ ├── predictor.py │ ├── random_forest.py │ ├── arima_model.py │ └── lstm_model.py │ ├── sentiment │ └── news_sentiment.py │ ├── data │ └── stock_data.csv │ ├── dashboard.py └── README.md Technologies Used Programming Python Libraries Streamlit Pandas NumPy Scikit-learn TensorFlow / Keras Statsmodels Matplotlib yfinance NLTK Tools VS Code GitHub Machine Learning Models Random Forest A supervised machine learning algorithm used for regression that predicts stock prices based on historical features such as: Open price High price Low price Volume ARIMA A time-series forecasting model used for analyzing and predicting future stock prices based on historical closing prices. LSTM A deep learning model capable of learning complex time-series patterns in stock price movements. Hybrid Model A weighted combination of multiple models: Final Prediction = 0.4 × LSTM + 0.3 × Random Forest + 0.3 × ARIMA This improves prediction accuracy by combining different modeling techniques. Installation 1. Clone the Repository git clone https://github.com/yourusername/stock-ai-project.git 2. Navigate to the Project Folder cd stock-ai-project 3. Create a Virtual Environment python -m venv venv 4. Activate the Environment Windows venv\Scripts\activate Mac/Linux source venv/bin/activate 5. Install Dependencies pip install -r requirements.txt Running the Dashboard Run the Streamlit dashboard using: streamlit run dashboard.py The application will open in your browser at: http://localhost:8501 Example Workflow Select a stock ticker from the sidebar Choose a prediction model View historical stock chart Click Run Prediction See predicted price and BUY / SELL signal Example Output Stock: AAPL Current Price: ₹15,842.21 Predicted Price: ₹15,210.10 Expected Change: -3.98% Signal: SELL Future Improvements Real-time stock price streaming Advanced technical indicators (RSI, MACD) Sentiment analysis from financial news Portfolio recommendation system Deployment to cloud platforms Author Durvesh Chaudhari MSc Data Science and Big Data Analytics License This project is developed for educational and research purposes.