Random Forest for Precise Predictions of Customer Experience at Restaurant X

Abstract

This study investigates the application of the Random Forest algorithm to predict customer satisfaction at Restaurant X, leveraging a dataset of 524 entries that include attributes such as service quality, cleanliness, food quality, and overall satisfaction levels. The research methodology comprises data preprocessing, exploratory data analysis, Random Forest model development, and evaluation using performance metrics such as accuracy, precision, recall, and F1-score. The Random Forest model demonstrated an overall accuracy of 72%, with its highest performance observed in the highly satisfied customer category, achieving an F1-score of 0.81. Analysis identified food quality as the most influential factor driving satisfaction, followed by service quality and cleanliness. However, the model encountered challenges in predicting dissatisfied customer categories due to class imbalance within the dataset. To address these issues, techniques such as Synthetic Minority Oversampling Technique (SMOTE) and additional data collection are recommended to improve model performance. This research underscores the potential of machine learning in providing actionable insights for the restaurant industry. Restaurant X can refine its operational strategies, address the root causes of dissatisfaction, and strengthen customer loyalty. This study demonstrates the capability of Random Forest to uncover critical satisfaction factors, enabling restaurants to optimize their service quality and customer experience.