Restaurant Data Analytics to Improve Business Verticals

ABSTRACT

Due to the Covid-19 pandemic, hospitality industry witnessed a massive decline in their revenues. In our research we realized that one of the most effective ways to aid customer retention and boost the revenue of this Our research shows that currently the data analysts in this industry only use the traditional tools for predictive analysis, offering from a limited range of offers that lack customization as per user purchase history. Hence, we put forward a proof of concept for a tool where we make a machine learning model that learns from the historic data of each restaurant, including customer segments and coupon parameters, and predicts the probability of a coupon to work on a specific sub-category of customers. This would thereby increase the chances of transaction and thus boost the revenue. We worked with several classification algorithms, like Logistic Regression, AdaBoost, Random Forest, Gradient Boosting, and realized that Random Forest Classifier was producing the best results. Thus we selected it for building our model. As a result, we have built a web-based tool that can be used by Analysts or the business person themselves, to find out what coupon offers would best suit a particular subset of customers. This would help them make better business decisions, gain more customer traction and retention, and consequently boost their revenue. © 2022 IEEE.