Sentiment Analysis of McDonald's Store Reviews in Fast Food Restaurants Using Machine Learning
Cong Doan Truong, Mai Xuan Dat, Nguyen Van Ninh · 2025
Sentiment analysis is important for understanding customer feedback, especially in the highly competitive fast-food industry, where consumer perception significantly influences brand loyalty and market positioning. This study evaluates the effectiveness of three machine learning models in analyzing sentiment from customer reviews of McDonald's stores across the United States, using the "McDonald's Store Reviews" dataset. The three models tested include Support Vector Machines (SVM), Naive Bayes, and Long Short-Term Memory (LSTM), with three different approaches: using two data balancing techniques Synthetic Minority Over-sampling Technique (SMOTE) and Random Oversampling (ROS) as well as a baseline approach without any balancing methods. The results reveal that the SVM model, balanced with SMOTE, achieved the maximum accuracy of 94%, beating the other models. This study exhibits the usefulness of SVM in sentiment classification and underlines the practical value of sentiment analysis in maximizing marketing tactics. The insights from sentiment analysis can help fast-food chains like McDonald's enhance customer satisfaction, improve marketing campaigns, and drive business growth. This research adds to building machine learning applications for sentiment analysis, a strong tool for data-driven decision-making in the fast-food industry.