Predicting Cryptocurrency Value using Sentiment Analysis

Abstract

This paper cross validates thesis given by few authors on the impact of social media on cryptocurrency prices. Initially, the focus is on the Bitcoin, later on, a similar model can be used for other cryptocurrencies. Sentiment scores of tweets and news feeds are considered along with historical prices and its volume to predict prices. Experimental results show that there is not much impact of sentiment scores unless these scores are not biased to one particular class.