/yelp_regression_project

supervised learning skill path - build a machine learning model with Python

Primary LanguageJupyter Notebook

Project: Yelp Rating Regression Predictor

Yelp Dataset Terms of Use - https://www.codecademy.com/content-items/6465513928f4b4eb1886c2bae85c72cd

Next Steps

You have successfully built a linear regression model that predicts a restaurant's Yelp rating! As you have seen, it can be pretty hard to predict a rating like this even when we have a plethora of data. What other questions come to your mind when you see the data we have? What insights do you think could come from a different kind of analysis?

- failure rate and/or when they will close
- build shadow profiles on "fans" for deeper insights
- suggestions for types of restaurants to open in certain locations

Here are some ideas to ponder:

Can we predict the cuisine of a restaurant based on the users that review it?
What restaurants are similar to each other in ways besides cuisine? location
Are there different restaurant vibes, and what kind of restaurants fit these conceptions?  of the features in this notebook top three are maybe child friendly, wifi public, and bike parking friendly.  I think Yelp collecting on pet friendly is an emerging one that would be helpful for various reasons including "vibe"
How does social media status affect a restaurant's credibility and visibility? restaurants that consistently put content that has good interaction rates on at least one social channel I agree would be interesting to model