Sentiment Analysis of Shopee App Reviews on Google Play Store Using Machine Learning Models

Abstract

This study aims to analyze user reviews of the Shopee application on the Google Play Store using machine learning models to classify sentiments into positive, neutral, and negative categories. Data collection was carried out by web scraping, resulting in 4,000 review samples. The preprocessing steps included cleaning, tokenization, stopword removal, stemming, and TF-IDF vectorization. Three classification algorithms were tested, namely Naive Bayes, Support Vector Machine (SVM), and Neural Network. Evaluation results showed that the Naive Bayes model achieved the highest accuracy at 81.2%, followed by SVM at 78.6%, and Neural Network at 75%. Furthermore, a sentiment analysis dashboard was developed using Gradio and deployed through Hugging Face to facilitate marketing decision-making based on user perceptions. This research proves the effectiveness of machine learning in sentiment analysis and offers actionable insights to enhance customer satisfaction.