Analytical Study of Customer Review Sentiment and Education Classification as the Basis for Customer Service Management in the Blu BCA Application

This study aims to develop a sentiment and education classification model to analyze customer reviews of the Blu BCA mobile banking application in order to identify positive and negative sentiments as a basis for improving customer service management and service quality. The study employs machine learning techniques using the Naive Bayes and Support Vector Machine (SVM) algorithms. Customer review data were collected through web scraping from the Google Play Store and processed using data cleaning, tokenization, stemming, and TF-IDF weighting. Model training and evaluation were conducted using Python with the scikit-learn library. The results indicate that both Naive Bayes and SVM models are capable of effectively classifying Indonesian-language customer reviews into positive and negative sentiment categories. Performance evaluation using confusion matrix, precision, recall, and F1-score reveals differences in accuracy and effectiveness in handling linguistic characteristics of customer feedback. This study can be applied in digital banking services, customer relationship management, service quality evaluation, and financial technology analytics. The novelty of this study lies in its focused application of machine learning-based sentiment analysis on Indonesian-language reviews of a digital banking application, integrating sentiment and education classification with customer service improvement strategies to transform large-scale customer feedback into actionable insights.