Customer Lifetime Value Prediction Using Machine Learning

Food delivery applications such as Zomato,Swiggy, and Uber Eats have transformed the dining experience by offering users access to a wide range of restaurants and cuisines at their fingertips. Despite their popularity, these platforms primarily focus on transactional convenience and lack deeper personalization, especially when it comes to understanding a user’s emotional state. Often, users struggle with the simple yet frustrating question: What should I eat right now? — a decision influenced not just by price or proximity but also by mood. This research presents MoodieFoodie, a novel food delivery web application designed to recommend food items based on the user’s mood. By combining multiple parameters — user ratings, restaurant cost, and distance — with mood-based suggestions powered by machine learning, MoodieFoodie delivers a uniquely personalized experience. It also opens its platform to small-scale vendors such as tea stalls and home-based sellers, broadening the range of offerings beyond conventional restaurant menus. The primary goal of this project is to develop an emotionally aware recommendation system that learns from user feedback, improves over time, and supports both customers and small businesses. This paper explores the system design, machine learning approach, experimental setup, and potential impact of such mood-driven personalization on the food delivery market. The system further supports small- scale vendors, expanding its marketplace beyond standard restraurants.