Inspiration
According to the latest research results, food production contributes over 30% of greenhouse gas emissions. Therefore, it is crucial to make people aware of the carbon footprint generated by the products they use every day to prepare their meals. Our application provides a quick and simple tool to evaluate an estimate of this number for a specific dish.
What it does
The app takes the name of the dish provided by the user and searches the Internet for a relevant recipe. Then, it finds the ingredients of the recipe and using NLP extracts the information about the amount and type of every single element used, evaluating the carbon footprint generated during its production. Finally, it displays the total "cost" and the contribution of each ingredient.
How we built it
Most of the code was written in Python, using:
- Selenium and BeautifulSoup for Internet scraping
- Gensim for word-data processing
- Flask for creating the visual aspects of the app
What we've learned
That shrimp and beef production uses enormous amounts of CO_2
What's next for FoodprintCheck
The app has considerable potential for future improvements:
- extending our searches to more cooking websites / calculating directly from ingredients or link provided by the user
- suggesting different versions of a dish to let the user choose the most relevant one
- recommending substitutions for the most impactful ingredients that will preserve the essence of the dish (our attempts to achieve this failed, as the system always recommended swapping meat for vegetarian options 😡)
Log in or sign up for Devpost to join the conversation.