Anomaly Detection in Cryptocurrency Prices

This Paper focuses on creating and utilizing ML algorithms that can detect anomalies in the pricing of key cryptocurrencies, including Bitcoin, Ethereum, Avalanche, Tether, and Litecoin. The volatility of the cryptocurrency market highlights the significance of identifying abnormal price movements that might indicate market scams or other discrepancies. Through this research, we employ Random Forest, Isolation Forest, and XGBoost models, all offering distinct advantages: Random Forest merges different educational approaches, Isolation Forest is skilled at pinpointing anomalies, and XGBoost applies enhancement techniques. By utilizing these models, we enhance the detection of abnormal trading behaviors, promoting a more secure and open environment in the cryptocurrency market. Additionally, we present the findings of the anomaly detection through PowerBI and Python tools to clearly communicate insights to stakeholders. The outcomes show the effectiveness of these machine learning approaches in detecting anomalies, aiding both investors and regulatory bodies in making well-informed decisions about the direction of the cryptocurrency market.