A location based social media / chat application
This is a chat application that demonstrates that location can be further integrated into our everyday applications.
Many chat apps exist but we still need to create our own groups and keep sending invites. Many people who send requests to join groups aren’t even near the event which can affect the quality of the conversation. This solution uses the technologies available to limit access to conversations to people who are close to the event. For example, imagine not having to start a slack channel or Fb group and keep asking people to accept invites for every event. People gather, a group forms, it gather places and events data, that’s it. Anyone in the hackathon can login!
For the purposes of this hackathon, the Geofencing API wasn’t used as it is hard to get a good signal indoors. However, the functionality that lets the device gather locations and put people into conversations nearby is implemented. Data is stored in key, value pairs on Firebase. The pipeline extends beyond this app, K-means clustering was used to gather the 3 clusters of people using Matlab.
Further functionalities can include a better implementation of the distance calculation using other API’s. Enhancing the pipeline by doing the clustering as a cloud function. Using a graph database for better semantics. Semantic analysis for a recommendation system.