Restaurant Rating Predictor

Projects | 2018 | Links:

Predicts a venue’s popularity solely from information available at the venue’s opening.

Note: The code and the end result cannot be publicly displayed due to copywrite by The Data Incubator; however, they can be sent privately upon request.

Using a combination of Linear Regression, Ridge estimation, and Random Forest Regression, I trained seperate models to predict the star ratings of restaurants based on their latitude and longitude, city, category, and attributes. After completing those models, I combined them into a more accurate full model using a FeatureUnion.

Learning Points: Transforms, Pipelines, Scikitlearn, FeatureUnions