Sentiment Analysis of Burger King Indonesia App Review using Latent Dirichlet Allocation

Published: 13 November 2025| Version 3 | DOI: 10.17632/2r533d2rhs.3
Contributors:
Jonathan Gilbert Sukmana, Hafizh Dzulqa Deiza, Mulyani Karmagatri

Description

This study aims to provide an analysis of consumer reviews of the Burger King Indonesia app. Technological advances have encouraged fast food restaurants to develop online delivery apps. This study was analyzed using data mining methods, RapidMiner, sentiment analysis, and LDA topic modeling to identify themes and topics that emerge from consumer reviews. This study also uses the CRISP-DM (Cross-Industry Standard Process for Data Mining) method to analyze data and manage data mining. The use of sentiment analysis helps in processing reviews, while the Latent Dirichlet Allocation (LDA) algorithm and RapidMiner can identify topics that will emerge in consumer reviews. The main results obtained in this study identified six main themes, namely: ease of service, customer experience, brand image, varied menu options, product quality, and applications. A total of 1066 reviews were collected from 2020 to 2024. The identified topics can be improved by enhancing the application service, improving service to maintain customer experience, building a strong brand image, creating new innovations in products, and improving the Burger King Indonesia application.

Files

Institutions

  • Bina Nusantara University

Categories

Computer Science, Business, Marketing, Data Science, Natural Language Processing, Machine Learning, Text Mining, Sentiment Analysis

Licence