PalateMap

Thumbnail
Logo
User Interface #1 (View Breakfast Lisitings)
User Interface #2 (Add New Breakfast Listing)

Inspiration

Our inspiration for PalateMap stemmed from the desire to make dining experiences more personalized and enjoyable. We noticed that while food delivery apps and restaurant menus offer a wide variety of choices, they lack the ability to tailor recommendations based on individual taste preferences. By leveraging AI and data science, we wanted to create a solution that could understand and predict what users would enjoy, making every meal a delight.

What it does

PalateMap analyzes the taste profiles of various dishes and creates personalized food recommendations. Users log dishes they’ve tried and rate the flavors, and the app uses this data to suggest new menu items that align with their unique palate. The system employs vectorization of taste profiles to rank dishes based on compatibility with the user's preferences, providing a tailored dining experience.

How we built it

Front-end and Back-end: Developed with React for a responsive and intuitive user interface. Integrated APIs like OCR to fetch restaurant menus. Data Processing: Leveraged Edamam's Food Database API to obtain detailed recipes for dishes. Used the OpenAI API to evaluate flavor parameters and represent them in vector format.

Challenges we ran into

Data Normalization: Finding databases which contained a variety of updated dishes and accessible information also proved to be difficult. Vector Representation: Accurately representing complex taste profiles as vectors required extensive tweaking and fine-tuning of our algorithms. User Feedback Integration: Developing a seamless way for users to log and rate dishes, and ensuring this feedback accurately updated their taste profile.

Accomplishments that we're proud of

High Accuracy: Achieving almost 100% accuracy in our taste profile predictions, as well as utilizing vectorization to represent dishes efficiently. Comprehensive Dataset Integration: Successfully integrating the OCR, Edamam and OpenAI APIs to provide a wide range of dishes and detailed taste profiles.

What we learned

Importance of Data Quality: High-quality and diverse data are crucial for training accurate models. The integration of multiple data sources and doing in depth research until we found the perfect API for our needs is vital for refining the recommendation system.

What's next for PalateMap

Mobile App Development: Create a mobile app to make it easier for users to log dishes and receive recommendations on the go. Restaurant Partnerships: Partner with restaurants to offer exclusive recommendations and deals based on users’ taste profiles.

By continuously improving PalateMap, we aim to revolutionize the dining experience, making every meal a personalized culinary journey.