Analisis Sentimen Opini Masyarakat Terhadap Vaksinasi Booster COVID-19 Dengan Perbandingan Metode Naive Bayes, Decision Tree dan SVM

Abstract

The impact of the COVID-19 pandemic is very broad, especially in Indonesia. In early 2022, Indonesia entered the early stages of recovering conditions caused by the pandemic. The government has an option for the community to carry out a third dose of vaccination (booster). However, there are a number of pros and cons in the community regarding the booster vaccine. This study aims to conduct sentiment analysis related to public opinion on the COVID-19 booster vaccination in Indonesia with Naive Bayes model. The data source used comes from Twitter. The workflow of this research includes data crawling, labeling, preprocessing, dataset sharing, and model testing and comparison with other models, namely Decision Tree and SVM. The results of this study indicate that the largest AUC score falls to the SVM model (75.40%), but for more accurate precision falls to the Naive Bayes model (83.81%). In addition, there is a confusion matrix which shows that the Naive Bayes model trial is running well.

Sentiment Analysis of Towards Electric Cars using Naive Bayes Classifier and Support Vector Machine Algorithm

“…The same research was also done by Kristiyanti et al (2018) who conducted research on the topic of comparing the Naïve Bayes and SVM algorithms for sentiment analysis of candidates for governor of West Java, the results of this study Naïve Bayes algorithm obtained 94% accuracy as the best accuracy [20]. Other research was also conducted by Aldisa and Maulana (2022) with the topic of comparing the Naïve Bayes, Decision Tree and SVM algorithms for analyzing the sentiment of the Indonesian people towards the Covid-19 booster vaccination, the results of the Naïve Bayes algorithm were superior to the Decision Tree and SVM algorithms with 83.81% accuracy [21]. The latest research was conducted by Undamayanti et al (2022) with the topic of applying the Naïve Bayes algorithm with Particle Swarm Optimization optimizer for analyzing student sentiment towards the MBKM program, the results of the Naïve Bayes algorithm modeling PSO optimization obtained an accuracy of 71.…”

Sentiment Analysis in Determining Service Performance between In-Driver and Gojek based on Public Opinion using the Naive Bayes Method

“…Tahapan dalam proses perhitungan klasifikasi kelas pada data uji dimulai dari perhitungan nilai prior probability, conditional probability, dan posterior probability. Berikut disajikan tahap perhitungan metode Naïve Bayes Classifier terhadap data uji tersebut [17]…”

Analisis Sentimen Masyarakat Terhadap Paylater Menggunakan Metode Naive Bayes Classifier

“…Naive Bayes juga diyakini merupakan metode untuk melakukan pemisahan data terstruktur yang lebih unggul daripada metode pemisahan data terstruktur lainnya dalam hal akurasi dan komputasi [15]. Pada penelitian oleh Aldisa dan Maulana membandingkan antara algoritma Naive Bayes, Decision Tree, dan SVM menunjukkan bahwa presisi Naive Bayes menempati urutan terbaik [16]. Penelitian lain yang dilakukan oleh Hasan dan Dwijayanti menunjukkan bahwa akurasi dari algoritma Naive Bayes sangat baik yaitu 92,5% [15].…”