-
-
Results Page
Inspiration
Do you have a rare, expensive, ingredient that just sits in your pantry for months? You only used it for one recipe, and afterwards it felt like such a waste.
Rare Recipify is here to help!
With the ease of pasting your recipe URL, we will help you find more recipes, so you can enjoy good food and keep your pantry fresh.
What it does
To enjoy the application, follow these simple steps:
🔗 Find a URL that holds your latest recipe! 🔎 Paste the URL into the search bar of our web application 🥬 (Optional) If you have dietary restrictions, select them now 🥘 Enjoy a list of tasty new recipes that share similar uncommon ingredients
How we built it
Rare Recipify uses Flask to support the backend, HTML templates and Flask for the front end, and NLP-python to clean text.
The backend was built by taking the recipe and scraping for the text that contains the ingredients. Next, we used NLP- python to clean text, looking for solely the name of edible foods. Finally, we take these ingredients and the user inputted health query, to find other similar recipes. We return these recipes name, url, and ingredients in JSON format.
The front end takes the JSON data, and formats it into a more visually pleasing manner. The recipe name is big and bold, ingredients are listed in perfect bullet points, and a green button to explore the URL of the recipe. We used HTML templates (Jinja), focusing on a pastel theme.
Challenges we ran into
At first, we struggled to do our styling with HTML templates and CSS files, but due to loading issues, we switched to using HTML templates with integrated files.
Next, the format of the websites we were scraping were inconsistent. We fixed this by using NLP to clean and sort ingredients. Due to many dependency and loading issues with NLP, returning the results is often very slow.
Finally, we struggled with dependency issues on our virtual environment. We fixed this by restarting the environment and redownloading all dependencies.
Accomplishments that we're proud of
The accomplishments that we are proud of including NLP texting cleaning, web scraping, and using scraped ingredients to find more recipes.
We are proud of using NLP to find other similar uncommon ingredients, and using those to find more ingredients. We have clean, well styled code, that communicates with the web efficiency.
What we learned
We learned many skills, both technically and personally. Technically, we learned a new tech stack, new API calls, and new design templates. Additionally, we learned how to integrate NLP into a fun web app. Personally, we learned how to work as a team, and how to network with other teams.
Finally, the workshops taught us new concepts, Professor Rogaway taught a Radical CS class, which taught us a lot about power in the technology field.
What's next for RareRecipify
We would like to ensure it works with all recipe website, and strength our NLP analysis. Cheers to the next!
Log in or sign up for Devpost to join the conversation.