This project creates a visualization of restaurant ratings. It uses machine learning on a yelp dataset to create a visualization of Berkeley divided into regions. specifically, a k-means algorithm is used to find clusters with centroids. These regions are shaded a certain color depending on the predicted ration of close restaurants. Yellow is 5 stars, blue is 1.

This project was created for computer science sixty-one A at Cal.

#######USE INSTRUCTIONS#########

You can create a user by going to the USER folder and creating a new .dat file with your prefs on restaurants

Get a list of all the restaurants in this dataset by running: python3 recommend.py -r

Generate a visualization by running -u to select a user from the USER directory: python3 recommend.py -u one_cluster

You can get finner groupings by increasing the number of clusters with the -k option: python3 recommend.py -u likes_everything -k 3

Predict what rating a user would give to a restaurant even if they havn't visited it by using the -p option: python3 recommend.py -u likes_southside -k 5 -p

Filter based on categories by using the -q option: python3 recommend.py -u likes_expensive -k 2 -p -q Sandwiches