Naïve Bayes-Based Data Analysis on Google Play Store’s User Review of Bank Jago Application

This dataset contains user reviews of Bank Jago, a digital banking application developed by PT Bank Artos Indonesia Tbk, which provides savings, lending, and transaction services. The data was collected from publicly available user reviews on the Google Play Store. These reviews, written in Indonesian, reflect user experiences and opinions regarding the application's interface, security, and transaction features. The purpose of this dataset is to analyze user sentiment and identify the key aspects that influence user satisfaction with the digital banking service.

A total of 2,000 reviews were gathered through web scraping using Python. After conducting data cleaning which are removing ambiguous, irrelevant, and duplicate entries, a final set of 1,432 high-quality reviews was retained for analysis. The dataset was manually labeled into positive and negative sentiments and processed using the Naïve Bayes algorithm in RapidMiner.

Initial findings show that 48.7% of the reviews are positive, while 51.3% are negative. Commonly mentioned themes include ease of transactions, security of personal data and funds, verification process, and customer service responsiveness.

This dataset is valuable for researchers in the fields of sentiment analysis, fintech, and digital banking, particularly those focusing on service quality. It can also be utilized as training data for machine learning models in natural language processing (NLP) and classification tasks. By analyzing customer feedback, stakeholders in the digital banking industry can gain insights to improve application functionality, strengthen customer trust, and enhance overall user satisfaction.