A Sentiment Analysis Approach for the Cryptocurrency Market and Blockchain Technology Using Naïve Bayes, Support Vector Machine and Random Forest
Denisa Elena Bălă, Stelian Stancu · Proceeding Papers · 2022
Virtual currencies or cryptocurrencies are based on Blockchain technology, also known as distributed ledger technology.As of March 2022, there are already over 10k virtual coins, their number being continuously growing since 2013.This paper aims to extract the public sentiment expressed towards the cryptocurrency market and Blockchain technology, two topics widely debated in the last decade.Our research was based on the use of Twitter data, collected with the help of an API in the RStudio environment.To identify the sentiment associated with the 5,000 tweets collected, we used the Bing lexicon approach and three supervised learning algorithms.The three classifiers are the Naive Bayes classifier, a Support Vector Machine and Random Forest algorithm.The accuracy of these algorithms was analyzed through four metrics, finding that the Random Forest classifier proved to be the most accurate, while the SVM algorithm offers the weakest results in terms of classification.The sentiment analysis conducted with the help of the Bing lexicon indicated a predominantly positive sentiment of online users regarding the cryptocurrency market and Blockchain technology.The present paper is structured as follows.The first part highlights a brief introduction to the issue of text mining analysis, as well as the area of supervised learning.Subsequently, a revision of the specialized literature in the approached subject is highlighted, by referring to some pertinent studies in this direction.The methods used as well as the data involved in this study are described in the chapter dedicated to research methodology.The paper continues with the presentation of the main results of the research, as well as with the highlighting of the conclusions and of some future research directions.The study concludes with an exposition of bibliographic sources.