Stock Price Prediction

Summary: This project is a stock price prediction model using LSTM (Long Short-Term Memory) neural networks. It leverages time series data, including historical stock prices and sentiment analysis data, to make future stock price predictions. The model is designed to assist investors, traders, and financial analysts in making informed decisions based on data-driven insights. It incorporates cross-validation, metrics evaluation, and visualization of actual vs. predicted values for comprehensive analysis.

Key Features:

• LSTM-based stock price prediction model.
• Integration of time series data and sentiment analysis.
• Cross-validation for robust evaluation.
• Metrics calculation (RMSE, MSE, MAE, R-squared) for model assessment.
• Visualization of actual vs. predicted values.
• Percentage of Predictions within ±10%: 84.39%

Usage:

1. Prepare and preprocess your dataset.
2. Train and evaluate the LSTM model using provided code.
3. Analyze cross-validation results and testing metrics.
4. Visualize actual vs. predicted stock prices.
5. Make data-driven decisions in the financial domain.

Requirements:

• Python
• TensorFlow
• Scikit-learn
• Matplotlib