Inspiration
Originally the idea came to light from Anjali, it started off as a simple food recommendation extension, and then we had a better idea. An extension that recommends small local restaurants in your area at the top of your page, to combat chain dominance in SEO.
What it does
LocalPoach is going to follow you around the internet until you want food! Once you search anything food related, say "Indian food" or "pizza near me", the first thing on your page will be recommendations of local restaurants related to your search. These recommendations will never feature a large chain restaurant. From there, users who opt in for our recommendations can obtain points and cash back from their purchase as incentive.
How we built it
We started off by trying to figure out how to determine our recommendations, we weighed our API options for this and narrowed it down to the YelpFusionAPI. The YelpFusionAPI had extremely thorough documentation and provided all the capabilities we need to accomplish recommendations. Yelp itself is largely built of small businesses, and we found that when using the user's geographical location, the term they looked up, the hot_and_new attribute, and have the entries sorted by lowest review_count were all we needed for obtaining 5 recommendations. We wrote the call initially in node.js+axios, and it is represented in the getFive(term, loc).
Our physical chrome extension code was written in javascript and jquery, and we essentially needed to scrape the user's google searches, which are stored in the "title" element in our web activity. After each google search, we'd stringify its object, parse through and use Google's Natural Language API to extract meaning from each word. This allowed us to determine whether or not the search was food related, if it is, then we call the getFive() using the user's location and the phrase they looked up. This will return 5 recommendations, which is formatted at the top of the page.
Luke got the extension webpage set up using ejs and express, that way users can signup/in to our services to start building points! This will be the user's hub to manage their Local Poaching. Users can also view a map of all the small and grossing business near them, which is configured using google's maps api.
Challenges we ran into
We had an issue figuring out how we're going to invoke the YelpFusionAPI, originally we wrote it in axios+node.js, but then were having some compatibility issues oddly. Tried out ajax+jquery, and that was giving us a CORS error, so we went back to axios+node.js. Was a small hiccup
Accomplishments that we're proud of
-Proud to get something deployed -Proud we got services working -proud we all grew
What we learned
Logan: I learned how to use ajax and axios rather than native https calls using node. Anjali: Using NLP and Luke:
What's next for LocalPoach
-user profile services -more on other causes (black owned, lgbtq owned, etc.)
Log in or sign up for Devpost to join the conversation.