Improving customer experience for air travel, sleeping accommodations, and restaurant recommendations.

What it does

Using Delta's API, we determine the customer's landing location. By using the landing location, we scrape TripAdvisor's search in order to find the most popular events and locations in the area, and ask the users if they are attending any of them. Based on the landing location and attending events, we algorithmically calculate the optimal location for sleeping accommodations based on the minimum-distance to each location using Google Maps API. We then return Airbnb's search results for that location.

Furthermore, we use Delta's API to determine the food offered and what times they were offered. Then, we can mathematically predict when the customer should eat his/her next meal for recommendations.

How we built it

We used HTML/CSS for the front-end, Python for the backend/API, and Flask to link the front-end with back-end. We used Delta's APIs to calculate the landing location and landing time, then used that data to crawl the TripAdvisor search results. Then, we finally took the data and used Google Maps API to find the best minimum-distance geocode location, and converted it to a street address that is a valid search input for Airbnb.

Challenges we ran into

Using Flask to connect front-end and back-end was difficult, as it was the first time we worked with REST/HTTP requests. However, with mentorship and independent research, we found a great methodology.


This was the first time we worked with company APIs, Python, and Flask. We successfully worked out the requests/connections!

What we learned

We learned to connect a scalable, scripted back-end with a simple front-end, while learning to quickly pick up on working with many different APIs!

What's next for ezStay

We want to become accurate predictors of the customers lifestyle patterns. We will want to integrate a database (MySQL) in order to store where customers decide to eat after our recommendations for food (hopefully we can integrate with Yelp). Then, we have the opportunity to work the data with Azure services to apply ML and predict where customers want to eat.

Share this project: