This repo is to explore using a hybrid approach to build a personalized restaurant recommendation system prototype using Yelp’s dataset and LightFM package. We also compared its performance with pure collaborative filtering model and with different loss function implemented in the package.
Dataset source: https://www.yelp.com/dataset
Github repository of LightFM: https://github.com/lyst/lightfm
For code demo please follow Restaurant Recommender.ipynb to see results of the analysis.
For questions please contact [email protected]