Best Feature Selection using Correlation Analysis for Prediction of Bitcoin Transaction Count

Abstract

Cryptocurrency made on the basis of block-chain technology Bitcoin is drawing the attention of individuals, corporations, governments and financial institutions today. As the number of Bitcoin transactions increases over the past years, the scale of the Bitcoin market has been increasing day by day. Predicting the number of transactions contained in a Bitcoin block is important in a Bitcoin network. The aim of this paper is to propose a learning feature selection method for designing a machine learning model that predicts the number of transactions contained in the Bitcoin block by applying the machine learning algorithm. Selecting the appropriate feature to design a machine learning model is crucial things to the performance of the model. We apply correlation analysis to select the appropriate learning feature of the transaction count prediction model in the Bitcoin block and verify the validity of the proposed method through experiments.