Affectionately dubbed "The Bad Bot," this project is a cryptocurrency trading assistant with a twist. Inspired by the complexities and fast-paced nature of crypto markets, we aimed to create a bot that intentionally suggests unconventional trading decisions for fun while still delivering powerful data analysis and market insights. This is the most effective way to spend your money if you have too much and do not know what to do with it.
Through this project, we explored the world of financial data APIs, candlestick chart visualization, and machine learning for predictive analytics. We also gained a deeper understanding of integrating tools like ARIMA models and random forests for price forecasting while keeping user interactions engaging and entertaining.
Real-time data was fetched from a cryptocurrency API, and the bot’s quirky "bad" decision logic adds a playful layer to traditional trading tools. To incorporate Moving Averages (SMA & EMA), Volatility Analysis, RSI, Support and Resistance Levels, External Features, Portfolio Optimization, and Sentiment Analysis into a predictive model, we can combine these metrics into a unified framework that informs predictions
Managing large datasets efficiently while maintaining real-time responsiveness was a significant hurdle. Suppressing unnecessary warnings during visualization and ensuring data consistency were also areas that required focus. Moreover, balancing the bot's intentionally "bad" suggestions with accurate data visualization posed an interesting creative challenge.
Ultimately, "The Bad Bot" serves as both a humorous and insightful approach to understanding market trends, showing that even in the volatile world of crypto, you can trade with a smile.
Log in or sign up for Devpost to join the conversation.