🚀 Bitcoin Price Prediction using RNN and LSTM

📊 Overview

Bitcoin is the most popular cryptocurrency nowadays due to its impressive returns, but investors still need to monitor its market trends as prices vary dramatically. This repository explores Bitcoin historical data, analyzes price trends, and performs time series analysis using RNN and LSTM neural networks to predict Bitcoin prices for the next 30 days.

🎯 Features

• Comprehensive Data Analysis: Extensive exploration of Bitcoin historical data
• Dual Architecture: Implementation of both RNN and LSTM models
• Missing Data Handling: Automated data extraction using Selenium
• Time Series Forecasting: 30-day price prediction capabilities
• Rich Visualizations: Multiple charts and trend analysis
• Pre-trained Models: Ready-to-use trained models included

📁 Project Structure

Bitcoin_price-prediction-using-RNN-and-LSTM/
├── 📊 bitcoin_price_predict.ipynb    # Main analysis and prediction notebook
├── 🔍 extract_data.ipynb            # Data extraction using Selenium
├── 📈 bitcoinunix.csv               # Scraped missing data
├── 📄 bitcoincharts.txt             # Additional Bitcoin data
├── 🤖 model/                        # Trained models directory
│   ├── timestamp_priceRNN.h5        # Trained RNN model
│   └── timeseries_price_LSTM.h5     # Trained LSTM model
├── 🎨 img/                          # Visualization assets
│   ├── site.png
│   ├── rnn_back.png
│   ├── rnn.gif
│   └── lstm_*.gif
└── 📋 requirements.txt              # Project dependencies

🚀 Getting Started

Prerequisites

• Python 3.7 or higher
• pip package manager

Installation

1. Clone the repository

   git clone https://github.com/yourusername/Bitcoin_price-prediction-using-RNN-and-LSTM.git
   cd Bitcoin_price-prediction-using-RNN-and-LSTM

2. Install dependencies

   pip install -r requirements.txt

3. Launch Jupyter Notebook

   jupyter notebook

4. Run the analysis

   • Open bitcoin_price_predict.ipynb for main analysis and predictions
   • Open extract_data.ipynb for data extraction procedures

📊 Dataset

The dataset is sourced from Kaggle - Bitcoin Historical Data. Due to missing data entries, additional data extraction was performed using Selenium web scraping (see extract_data.ipynb).

Key Data Features:

• Historical Bitcoin prices (OHLCV data)
• Timestamp-based data points
• Over 4.8M+ data entries
• Missing data handled through web scraping

🤖 Models

RNN (Recurrent Neural Network)

• Architecture: Basic RNN layers with dropout
• Purpose: Baseline time series prediction
• Model File: model/timestamp_priceRNN.h5

LSTM (Long Short-Term Memory)

• Architecture: LSTM layers with regularization
• Purpose: Advanced time series prediction with memory
• Model File: model/timeseries_price_LSTM.h5

📈 Results

The models provide 30-day Bitcoin price predictions based on historical patterns and trends. Detailed analysis and visualizations are available in the main notebook.

👨‍💻 Author

Anuj Dev Singh
Machine Learning Engineer & AI Scientist

🛠️ Technologies Used

• Python 3.7+: Core programming language
• TensorFlow/Keras: Deep learning framework
• Pandas & NumPy: Data manipulation and analysis
• Matplotlib & Seaborn: Data visualization
• Scikit-learn: Machine learning utilities
• Selenium: Web scraping for missing data

📝 License

This project is licensed under the MIT License - see the LICENSE file for details.

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

📧 Contact

For questions or suggestions, please open an issue in this repository.

---

⭐ Star this repository if you found it helpful!