Restaurant Recommendation System Using User Based Collaborative Filtering

Abstract

A recommendation system is an application that can identify entities of interest for a person and provide suggestions based on the past record of person’s likes and preferences. The entity of interest can be anything, for example it can be a product, a movie or a news article. Recommender system is an effective way to help users to obtain the personalized and useful information. However, due to complexity and dynamic, the traditional recommender system cannot work well in mobile environment. Keeping such things into consideration, this recommendation system aims to recommend restaurants to users using their past preferences so they do not need to go through a list of choices. The recommender system adopts a user preference model by using the features of user's visited restaurants, and utilizes the location information of user via GPS(Global Positioning System) using LBS(Location Based System) and restaurants to dynamically generate the recommendation results using collaborative filtering technique. The suggestions will be based on the user preferences obtained from the past ratings and reviews given by the user, frequently visited cuisines of the user and the time preference of the user. Moreover, a brief analysis of reviews is also made to provide user a computed synopsis of the restaurant using text mining algorithm.