Sentiment Analysis of Seabank Application Reviews on Google Play Store Using Naïve Bayes Classifier

In the era of rapiddvancement in information and communication technology, the SeaBank application has attracted significant attention due to its attractive high interest rates on savings and deposits, ranging from 5% to 6%. This makes digital banking more attractive than conventional banks. The study utilized the Naïve Bayes Classifier method to analyze sentiment from a total of 820,000 reviews with a sample of 2492 user reviews, which were categorized into positive and negative, excluding neutral reviews. The results of the analysis showed that most of the reviews, about 54.7%, were negative. Words that often appear in positive reviews include “good,” “easy,” and “fast,” while negative reviews often highlight the words “disappointed,” “difficult,” and “failed.” The evaluation of the classification model shows that the Naïve Bayes Classifier has an accuracy of 93.81%, with an F1-score of 93.8%, a recall of 93.81%, and a precision of 93.82%. The conclusion of this study confirms the effectiveness of the Naïve Bayes Classifier in analyzing the sentiment of banking app reviews, providing valuable insights for the future development of SeaBank applications.