BITCOIN PRICE PREDICTION USING LASSO ALGORITHM

Abstract

In this paper, we proposed to predict the Bitcoin price accurately taking into consideration various parameters that affect the Bitcoin value. For the first moment of our paper we gathered the data set in Quandl site. Which come with some pure dataset. The data set consist of various features related to the Bitcoin price and trading value. Then we have done the training and testing for dataset. At the second stage of our paper, we have applied some prediction techniques as Linear Regression model to perform the data set very accurately for getting the MAE, MSE and r-squared values. Then we move on to the next stage as AI algorithm. In this paper, we have used the LASSO (Least Absolute Shrinkage Selection Operator) machine learning algorithm. This helps to do the selection operator in large dataset and performing a fabulous for finding the maximum accuracy. These two algorithms are very helpful for finding the Bitcoin price very high accuracy. In the Finally stage of our investigation, using the available information we tested the results by real-time results. Then we get the approximately accuracy then the Real-time accuracy. The sign of the daily price change with highest possible accuracy will be predicted.