People make decisions everyday. These decisions might be related to choosing which movie to watch or which city to travel to or which restaurant to eat in. Recommendation systems are an efficient way to help people make decisions in these situations. These systems do so by filtering all available information and presenting a list of items which are likely of user interest. Recommendations are based on certain characteristics. These characteristics may be from the user's social profile (Collaborative filtering approach) or from the item information (Content based approach). A plethora of different Recommendation systems have been created for various machine learning applications in numerous disciplines. For example, Amazon.com uses recommendation algorithms to personalise the online store for each customer. The store changes based on customer interests.

In our project, we have provided two kinds of recommendations. One is a city-based recommendation where the most popular restaurants in a city are recommended to the user based on the sum of checkins for the entire week. We have displayed the most popular restaurants for 11 cities. The second kind of recommendation is the personalised recommendation where we provide the user with a list of the most popular restaurants in a given city on a particular day. This recommendation is based on the number of checkins as well as the reviews for the restaurants.