Comparison of Machine Learning and Deep Learning in Shopee Review Sentiment Analysis

Abstract

This study aims to compare the performance of Machine Learning algorithms (Random Forest and Support Vector Machine) and a Deep Learning model (Long Short-Term Memory) in analyzing user review sentiment of the Shopee application. A total of 50,000 Indonesian-language reviews were collected through web scraping from the Google Play Store. After preprocessing and feature extraction, the three models were developed and evaluated using accuracy, precision, recall, and F1-score metrics. The results indicate that the LSTM model achieved the best performance in classifying sentiment into three categories: positive, negative, and neutral. Furthermore, the model was implemented into an interactive sentiment analysis dashboard using Streamlit, enabling users to explore and test sentiment in real time. This research demonstrates that the application of Machine Learning and Deep Learning technologies is effective in analyzing public opinion and can support strategic decision-making in the context of e-commerce.