Machine Learning based Bitcoin Price Prediction

Abstract

The decentralized digital currency known as Bitcoin has been used for trading and investment. To make wise judgments, traders and investors need to be able to predict the price of Bitcoin. In this paper, we attempted to predict the Bitcoin price using Machine Learning (ML) techniques. In particular, we utilized Long Short-Term Memory (LSTM), KNearest Neighbours (KNN), Random Forest Classifier (RFC) and Extreme Gradient Boosting (XGBoost) to predict the Bitcoin prices. These models are trained and evaluated using historical Bitcoin price data such as volume, open, high, and low prices. We have compared the performance of the ML algorithms with different parameters such as F1-score, recall, accuracy, and precision. The outcomes of the experiments show that the performance of the XGBoost model is better than other ML algorithms. The accuracy of the XGBoost model is more than 92%. The XGBoost model also obtained better results in terms of precision, recall, and F1-Score.