A real-time Bitcoin price prediction system using LSTM models and Twitter sentiment analysis.
Bitcoin Prediction is a machine learning system that forecasts Bitcoin prices using LSTM neural networks and Twitter sentiment analysis. It processes real-time market and social media data to generate buy/sell signals for cryptocurrency trading. The project provides a complete pipeline from data collection to visualization through Tableau dashboards.
Data scientists and developers interested in cryptocurrency trading algorithms, time-series forecasting, and sentiment analysis applications. Particularly useful for those exploring practical implementations of LSTM models for financial prediction.
It offers a working implementation that combines multiple data sources (price data and sentiment) for improved prediction accuracy. Unlike theoretical demonstrations, it provides a complete trading system with real-time data processing, model inference, and visualization components.
This is the code for "Bitcoin Prediction" by Siraj Raval on Youtube
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
Continuously fetches Bitcoin price and Twitter sentiment using Python scripts, storing data in CSV files for up-to-date predictions, as outlined in the data gathering section.
Uses LSTM neural networks that outperformed ARIMA models in experiments, providing better accuracy for time-series forecasting, as mentioned in the core engine description.
Incorporates Twitter sentiment analysis as a feature for price prediction, enhancing accuracy by combining market data with social media trends.
Generates buy/sell decisions based on predicted prices and configurable thresholds, making it actionable for trading applications, with output stored in a MySQL database.
Offers a full pipeline from data collection to visualization via Tableau, emphasizing practical use over theoretical exploration, as highlighted in the philosophy.
The entire setup works on Google Cloud, requiring cloud infrastructure and potentially incurring costs, limiting flexibility for local or budget-conscious deployments.
Relies on Tableau for real-time visualization, which is proprietary software and may not be accessible for all users, especially in open-source projects.
Credits go to an external project and it's a wrapper from a YouTube video, suggesting it might not be actively maintained or use current ML best practices.
As a wrapper, it may not allow easy modification of core components like the ML model or data sources without significant code changes.