Inspiration
Whenever we want to make meals, we don't always have all the ingredients, or we aren't able to include those foods in our diet because of allergens.
What it does
Suggests substitutes for ingredients in your recipe.
How I built it
Scrapping a recipes website for HTML tags and classes indicating ingredients and instructions. From there, I treated adjectives and verbs related to each ingredient as binary features in a sparse matrix. I then used the scikit learn library to try out different prediction models, and used the predict_proba function to check the other similar options to offer as substitions/alternatives to the ingredients in the user's recipe.
Challenges I ran into
Designing the data representation for each ingredient.
Accomplishments that I'm proud of
Scrapping and training the model.
What I learned
Language processing edge cases.
What's next for TuReceta
Android application to serve results of users.
Log in or sign up for Devpost to join the conversation.