We love to eat at food trucks. We wanted to make it easier to find food trucks that we were interested in while also promoting small businesses by giving them a smart platform to attract and engage their communities.
What it does
Vend_ is an application that utilizes AI and machine learning to recommend nearby vendors that users are likely to be interested in. Vend_ is currently an Android app that both users and vendors can log into and use. Vendors are able to log in and post their current location to our database. Users use the app to get push notifications for nearby vendors that may be of interest to them. When Vend_'s Algorithm detects that a user is likely to be interested in a nearby vendor, it sends out a push notification that recommends the vendor to the user and asks them if they plan on going to that vendor. If the user is interested in going to the vendor the app will display a map of the vendor's current location. All of the user's decisions help to feed our model of their preferred vendor type.
How we built it
The Vend_ app is built from Android studio. The back end of Vend_ runs on Firebase and Tensorflow. The google assistant action runs on the Google Cloud Platform.
Challenges we ran into
None of our members have had much experience developing Android apps. Two of our team members have never used Android studio and one has never even coded in Java. Quickly learning these new skills was a big challenge.
Accomplishments that we're proud of
We used machine learning to figure out which vendors to recommend to users and we used Google Maps to help locate said vendors.
What we learned
Two members of the team have not used Android Studio and learned enough to contribute.
What's next for Vend_
Make the UI prettier and make an iOS app. Also, work to improve the recommendation model.