For my final data science project at Metis, I wanted to make something that was creative, contributed to society, and that I was passionate about. I thought about what I like to do on the weekends, and one of my favorite things to do is to go to art museums, specifically, the Metropolitan Museum of Art in NYC.

Last summer, I visited the Met with a friend who didn't have much of an art history background. Although I know a little about art history, it was difficult for us to decide which galleries to visit while looking at the map because the map wing names and movement names aren't very informative of the content inside the galleries.

This was a challenge that Data Science could solve. I made a recommender system for paintings at the Met in which the user chooses the paintings that she likes, and the recommender finds the most similar paintings using a cosine similarity matrix and recommends the gallery numbers that you can find those paintings on display.

Stay tuned for the web app and mobile app!