And the Winner of the Michelin Restaurant Guide Challenge is…
After weeks of preparation, data crunchers flocked to campus today to present their predictive skills in the ultimate epicurean test: the DC Michelin Restaurant Guide Challenge. Three out of the top five contenders were GA DC Data Science Immersive students. The ultimate winner was Brendan Sudol, a developer at 18F and former data scientist at Etsy.
Sudol described his winning solution as a “Nate Silver-esque” approach: “I really don’t know too much about the food scene in DC, so I relied on averaging and aggregating other reviews.” Sudol determined there would be between seven and 20 restaurants on the Michelin Guide, given the ratio of Michelin Starred restaurants in other Michelin-reviewed cities cities relative to number of four-dollar-sign Yelp restaurants. Upon settling on a range, he averaged popular sources like Washingtonian and Washington Post’s Tom Sietsema reviews.
Sudol ultimately submitted 14 restaurants, correctly identifying seven and the number of stars those seven would earn. Further information about Brendan’s methods are available here. The official Michelin list is here.
To learn more about our data science and data analytics programs, classes, and events, check out our website.