Analisis Sentimen terhadap Data Ulasan Aplikasi Mobile Banking di Play Store

Abstract. Advances in information technology have driven the transformation of banking services, one of which is through mobile banking. This study aims to analyze sentiment toward the Bank Rakyat Indonesia (BRI) and Bank Syariah Indonesia (BSI) applications based on reviews on the Google Play Store, as well as to compare the performance of the K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Random Forest, and Naive Bayes methods. The data used consists of 6,000 reviews for each app, collected between January and July 2024. The analysis process includes data pre-processing, sentiment labeling, training and evaluating models using accuracy, precision, recall, and F1-score metrics. The results showed that Random Forest and Naive Bayes performed best, with the highest accuracy for the BRI app at 81% and BSI at 78%. Overall, BRI maintained a positive trend and demonstrated a strong digital reputation, while BSI still needs to continue improving its authentication system and service response to reduce negative sentiment. Abstrak. Perkembangan teknologi informasi telah mendorong transformasi layanan perbankan, salah satunya melalui mobile banking. Penelitian ini bertujuan untuk menganalisis sentimen pengguna terhadap aplikasi Bank Rakyat Indonesia (BRI) dan Bank Syariah Indonesia (BSI) berdasarkan ulasan di Google Play Store, serta membandingkan performa metode K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Random Forest, dan Naive Bayes. Data yang digunakan berupa 6.000 ulasan untuk masing-masing aplikasi, diambil pada periode Januari–Juli 2024. Proses analisis meliputi pre-processing data, labeling sentimen, melatih dan mengevaluasi model menggunakan metrik akurasi, precision, recall, dan F1-score.. Hasil menunjukkan bahwa Random Forest dan Naive Bayes memiliki kinerja terbaik, dengan akurasi tertinggi pada aplikasi BRI sebesar 81% dan BSI sebesar 78%. Secara keseluruhan, BRI berhasil mempertahankan tren positif dan menunjukkan reputasi digital yang kuat, sedangkan BSI masih perlu terus menyempurnakan sistem autentikasi dan respons layanan untuk mengurangi sentimen negatif.