Implementasi analisis sentimen pada ulasan aplikasi Duolingo di Google Playstore menggunakan algoritma Naïve Bayes

Abstrak

This study investigates sentiment analysis to evaluate the Duolingo Application using the Naive Bayes method. The Duolingo program exemplifies the use of big data technology for processing vast and complex data. Google Play Store offers review and rating functions that can help with program development and fixing undesirable aspects. This project uses sentiment analysis techniques to automatically analyze Indonesian internet product reviews and obtain information about the feelings expressed in these reviews. The Naïve Bayes method is used to classify reviews as positive or negative. The research findings show that a dataset consisting of 1000 pieces of data originating from reviews of the Duolingo program on the Google Play Store was manually labeled before the preprocessing step. Of this number, 500 data have positive sentiments, while 500 have negative attitudes. Additionally, sentiment analysis shows an accuracy rate of 86%. The f1 score precision is 89% and recall 83% values, with f1 results for classification of 86%.