Recommending Businesses on Yelp

This project looks at and evaulates some different recommendation strategies for matching users with business on Yelp

Summary

In this project I created a Neo4j graph database on AWS to store data from the Yelp Challege dataset. I then used python and the cypher query language to make a simple graph based recomendation algorith and compare the performance of that algorithm to some standard recmmendation models. I used the Surprise python package to train and test the standard models.

A copy of my slideshow presentation is in this repo with the name Yelp Recommendation

The tech demo that goes along with this project is in this repo