[Week 1 — GourmetNet]
Introduction
With GourmetNet our goal is to develop a food recommendation system for Yelp. There are many examples of recommendation systems but the most prominent one we think is on Amazon. There are many approaches to this but we preferred the baseline method the collaborative filtering.
On the project we are going to share our code along with explanations. Following this you can implement your own recommendation system.
What Did We Do?
We started to look how other recommendation systems are done and to our luck there are many examples. But most of them use the collaborative filtering algorithm.This problem seems like mostly figured out in many ways. We are just going to implement the baseline method and see if we can improve it. We decided to use public dataset from Yelp to build our model.
Our First Challenge
The first main problem we faced was how sparse the matrix of our data. It is really hard to analyze the similarities just from the certain restaurants alone. It is hard for two people to write a review for the same restaurant let alone many restaurants. So we thought to overcome this we need to group our data. We are going to spend our next week trying to achieve this.
That’s it for this week. See you next week! By the way here is the dataset we are using.
