FeastFinderAI

A spatial analysis project on NYC restaurant inspection data. It trains a model to predict restaurant grades and uses that to suggest areas by postal code, then plots the recommendations on a map.

Approach

• Data: DOHMH New York City Restaurant Inspection Results, using name, zip code, grade, score and coordinates.
• Model: scikit-learn RandomForestClassifier on the inspection fields.
• Output: recommended restaurants for a given postal code, drawn on a Folium map.

Running it

pip install pandas scikit-learn folium

Download the dataset, point the file path in the notebook (predictive_dining_NY.ipynb) at it, and run the cells.

Only the README and license are currently in this repository; the notebook and script were not carried over, so this is documentation of the approach rather than runnable code.