What's Cooking? Predicting Cuisines from Recipe Ingredients

Abstract

Kaggle is an online platform for hosting data science competitions. On September 9, Kaggle opened a contest called ``What's Cooking?''. The goal of this contest is to predict the type of a cuisine (e.g. ``Chinese'' or ``Mexican'') from the list of ingredients in a recipe. In this paper, we describe our methodology for training an algorithm to make such predictions, which ultimately achieved a score of .78107 on Kaggle, placing 366th out of 841 teams.

Writeup

http://kevinkdo.com/sta561/final_project.pdf

Code

https://github.com/kevinkdo/sta561final

Poster

http://kevinkdo.com/sta561/main.pdf