Subject/s of Focus: Big Data and Cloud Computing

Authors: James Labergas, Chelsea Ong, Lance Sy, Maisie Tan

My team developed a collaborative-based recommender system via Alternating Least Squares and a content-based recommender system using restaurant reviews scraped from the Zomato website. The recommendations generated can be used by restaurant owners and customers in maximizing the platform.

This project is very memorable for me because I presented for the first time in public. It was nerve wracking but overall it was fulfilling.

This project is for the fulfillment of the requirements in our Big Data and Cloud Computing course.

Abstract

Launched in India back in 2008, Zomato operates with the vision of bringing better food to people. To date, it is one of the largest food aggregators with operations spanning across 24 countries. With the millions of users on its platform and the customer feedbacks available in their page, there is no doubt that the platform has a rich understanding on its customers’ respective food profiles. With the objective of generating more value for the customers and other Zomate stakeholders, this project aims to address this question, Can we develop a personalized restaurant recommender system for Zomato users?. Using restaurant reviews data scraped from the Zomato website, we develop a collaborative-based recommender system via Alternating Least Squares, as well as a content-based recommender system. With these newly developed recommender systems, we then make recommendations to restaurant owners and customers to maximize the Zomato platform in understanding their customers more and to gain more traction for their food service establishments. We also propose for future researchers to include more filtering options and introduce more countries into the study.

Note:

For more information on the files and codes used for this project, you may send me a message via Linked in