Classification Algorithm for the Sentiment Analysis of the Covid 19 Pandemic in Indonesia Context
Keywords:
Covid 19, twitter, Naïve Bayes algorithm, Sentiment analysis, TextblobAbstract
The Covid-19 pandemic that is happening in Indonesia has had a significant impact both in terms of economy, education and health. Pros and cons related to the policies made by the government in tackling the impact of the spread of the virus produces a variety of sentiments in the community that is interesting to analyze. By taking sentiment data on Twitter, which extracted using Tweepy, then the tweets data are analyzed with Textblob library contained in Python. Naïve Bayes algorithm then, have been used to classify the tweets either in positive, negative or neutral sentiments. From the experiments conducted, the test results found that there were 60% positive sentiments, 15% negative sentiments and 25% neutral sentiments on economic topics. On the other hand, 70% positive sentiment, 3% negative sentiment and 27% neutral sentiment on the topic of health. Meanwhile, 10% positive sentiment, 50% negative sentiment and 40% neutral sentiment on the topic of education.