You spent $10 buying a box of spinach from Wholefoods and decided to live a healthier life. However, you totally forgot this ingredient every day. One week later, the $10 organic spinach rots. YOU HAVE TO THROW THAT AWAY. You have been eating burgers and pizzas for the whole week! So, we were thinking, why can’t we build a website that helps reduce food waste and leads people to a healthier life!

What it does

The user could upload pictures, even handwritten ones(Test for AWS Rekognition API), to the website to be processed into ingredients with expiration dates, or doing checkboxes to identify wanted ingredients and the website will generate best recipes based on algorithms.

How we built it

We got our data from web scraping with python. Javascript takes input as a string and sent to the python algorithm to concatenate the URL to complete the web scraping. Using a pivot table, we create unique mappings for future searches. With a great amount of data, our project is able to recommend the user the most suitable recipe. To input the ingredients data, we provide a method using aws image and text recognition to allow the auto-generate ingredient list that will expire soonest. Accessing the database we built with AWS services and do queries, we have a webserver to do server-side algorithms including image recognizing(AWS Rekognition)and HTTP methods in javascript, and frontend web app for the user to interact with.

Challenges we ran into

This is our first time ever that we fully utilizing the full stack and create the interactive web app with full database, express and Apax connected client-side and server-side and implementing animations with CSS. We have learned AWS from scratch and we had some trouble setting up the server. Setting up Rekognition API in Node.js and format and process the data generated by the API is another challenge. We also ran into issues in establishing connections among clients and the webserver.

Accomplishments that we're proud of

We built a functional dynamic website, with AWS Rekognition API implemented and stunning visuals in less than 2 days.

What we learned

How relational databases work in a real web application, AWS, SQL, image recognition, web scraping, teamwork, communication

What's next for Recipe101

We want to improve our recommendation system and maybe make it mobile as well.

Share this project: