A Classification Model for Prediction of the Cafe Prospective Customer

The café business faces declining customer numbers and sales due to increased competition serving diverse customer preferences. Understanding target customers and defining key characteristics that attract them is essential. This research investigates the factors influencing customer decisions to choose a café in Chiangrai, Thailand particularly focusing on preferences for cafes that offer a calm, natural, outdoor, or chillout nightlife environment. Through surveys and data analytics, we identified key factors influencing café choices and predicted the likelihood of customers selecting a café based on their preferences. We surveyed local residents, dividing them into two groups: those who frequented this type of café and those who frequented other types. This paper proposes a machine learning classification model to predict whether a customer is likely to choose this café style or not. The model aims to accurately classify potential customers based on their likelihood to choose a particular coffee shop. Machine learning models, including Random Forest, Decision Tree, and Naïve Bayes, were compared for predicting target customer choices. To improve their performance, hyperparameter tuning was applied. Experimental results showed that Random Forest achieved the best performance across all measures, reaching 94.38% accuracy, 94.40% precision, 94.40% recall, and 94.40% F1-score. These findings provide valuable insights for cafes to improve services, attract the right customers, and create effective marketing campaigns by customizing their offerings and marketing to specific customer groups.