Inspiration

The holidays are a time when people gather together, rejoice and make memories. Sharing a meal, laughter and dessert with your favorite people in the world is one of the most fulfilling experiences and food is ultimately what helps bring people together. Unfortunately due to the pandemic, getting your favorite foods while staying safe and socially distanced is challenging. This sucks for the two of us as we are really big Foodies.

Therefore, we built this platform to help people physically distance, get the food they enjoy, and help local restaurants. We hope this app helps keep people safe during the holiday season and maybe even motivates people to explore new places to eat and support small businesses.

What it does

Our platform has two goals. First to help people find their favorite foods, while still remaining safe and socially distanced during this Holiday season and second to bring some cheer to the local restaurant owners that have also been affected during this pandemic.

How it works:

  1. Identifies all the restaurants in a nearby chosen location (allowed to sort by cuisine of interest + other filters)
  2. Using a user-developed algorithm, predicts the relative density of all the restaurants in the vicinity compared to its average crowdedness. Any restaurants that are not currently very active are labeled as "Santa Approved"
  3. Identifies if the nearby area is crowded/overwhelmingly dense using geofencing and informs the user to be cautions.
  4. Using a Web Crawler, also identifies the relative crowdedness of each restaurant during the week (since it is not available with google API)

How we built it

There were a lot of different aspects to the technical details

  • The application runs with NodeJS and is currently hosted on Heroku.
  • Google Places API allows us to call for all the nearby search restaurants in a nearby location
  • However, the Google API does not yet support population information, which we gathered based on weekly visits and a Google Search API call alongside a web crawler parsing through each search result.
  • Also, many locations are missing live population information, so we wrote an algorithm to do predict these values using Google Cloud from weekly crowd information extracted from the web crawler
  • To Measure the Relative Safety of a given location, Radar.io was used, measuring the foot distance to all nearby restaurants that are not "Santa Approved"
  • ChartsJs was used on the frontend to visualize all the population patterns.
  • Designs were edited in Gimp and Made with Hatchful by Shopify

Every time the user enters a location, the Google Places identifies any near by restaurants. We feed the address of each restaurant to the web crawler + the developed algorithm, which are able to produce the weekly population patters at that location, along with a current population prediction This information is then fed into a geofencing function, which uses Radar.io to identify if the currently location is in proximity to dense areas, allowing the user to pinpoints socially-distanced areas

Check out our Github Below or Just Let us Know if you want more technical specifications

Share this project:

Updates