parikshitjanagama / yelp-recommneder-system

• Developed a Recommender System for restaurants by performing analysis on data preprocessed from Yelp Dataset. • Used Altering Least Squares method with Matrix Factorization and Neighborhood Model to train and build the Recommender System. • Tested the Recommender System with multiple rounds of Cross Validation technique and 16% prediction error is observed

Yelp-Recommneder-System

This project repo consists of Yelp Recommender System for Restaurants. Used two methods Neighborhood Model and Latent factor model of Collaborative Filtering for our Recommender System. Root mean squared error(RMSE) metric is used to see know how close the prediction ratings are to the true ratings Planned to split the data on 80:20 ratio for Training and Test Data and run 5 iteration RMSE. Spark Programming using Scala and Java are used to build the system.