Recommendation System for Restaurants

Abstract

Managing a restaurant can be a competitive and an arduous business. According to the Business Insider, as many as 60% of restaurants close within the 1st year of opening [1]. This competitive nature of the restaurant business makes it imperative for owners and managers to understand their customers at a more personal level. Predicting user preferences and recommending food items and restaurants are starting to become a popular advantage. Apart from driving up sales and creating a more personalised experience for the user, it drastically speeds up searches making it very convenient for the customer. In this project, we wish to propose a hybrid recommendation engine that bridges the gap between customers and restaurant owners by aiming to ameliorate user experience and increase restaurant sales. We have built a recommendation engine that enables restaurants to enhance foot-fall and give personalized recommendations to users by focusing on reviews, psychographic factors, lifestyle patterns and user location. By making use of methods like content-based filtering, collaborative filtering and using user's Google check-in data to identify locations, we have created a hybrid recommendation system. We present a hybrid recommendation engine, that aims to alleviate problems present in each of the methods and create a foolproof robust system.