Sentiment Analysis on the Pros and Cons of Cryptocurrencies using the Multinomial Naïve Bayes Algorithm

Public awareness of the existence of cryptocurrencies in Indonesia is increasing. However, there are not a few people who reject cryptocurrencies, there are still many Indonesian people who refuse for various reasons, one of which is because the very high price fluctuations of this currency make inv...

Full description

Saved in:
Bibliographic Details
Published in2023 IEEE 9th International Conference on Computing, Engineering and Design (ICCED) pp. 1 - 6
Main Authors Gunawan, Rafli Indra, Wahana, Agung, Maylawati, Dian Sa'Adillah, Slamet, Cepy, Alam, Cecep Nurul, Arianti, Nunik Destria
Format Conference Proceeding
LanguageEnglish
Published IEEE 07.11.2023
Subjects
Classification algorithms
Cryptocurrency
multinomial naïve bayes
natural language processing
Sentiment analysis
Social networking (online)
Training data
twitter
Vocabulary
Waste materials
Online AccessGet full text

Cover

More Information
Summary:Public awareness of the existence of cryptocurrencies in Indonesia is increasing. However, there are not a few people who reject cryptocurrencies, there are still many Indonesian people who refuse for various reasons, one of which is because the very high price fluctuations of this currency make investing in cryptocurrencies an investment with high risk. Therefore, this research aims to find out public opinion on cryptocurrencies through the comments of Twitter social media, including positive or negative sentiments. This research classifies tweets with positive and negative opinions on cryptocurrencies using the Multinomial Naive Bayes (MNB) method. The research experiment results show good performance with an accuracy of 79%, a precision of 78% and a recall of 85%. Then, the in-depth sentiment analysis results indicate that most Indonesian people have more positive sentiment (71.7%) on cryptocurrencies issues than negative sentiment (28.3%).
ISSN:2767-7826
DOI:10.1109/ICCED60214.2023.10425409