When harvey came most of the information we were getting was form different Facebook posts and twitter tweets. A lot of information, call for help was going out through these channels. And much of them were unmonitored for obvious reasons. When i myself was in trouble I was constantly trying to monitor my Facebook and twitter feed to see what is happening around me.
Now this app can do all that and also track all the call for help based on severity of the situation as well.
What it does
It monitors Facebook, Twitter, website rss feeds and specific sms channels for any specific keyword configured and prioritizes base don that. It can generate reports in real-time, can process 1000 reports per second from realtime twitter and facebook feeds combined also can group based on Geolocation of the incidents. It has helpful option on assigning any incident to specific responders and group incidents as well for easy monitoring and responding.
How we built it
HarveyTrack consists of two largely separate frontend and backend apps. Some model code (in /shared) is shared between them.
The backend is a Node.js/Express app responsible for fetching and analyzing data and servicing API requests. There are three main modules, each of which runs in its own process:
API module Fetching module Analytics module
The model layer (in /models) is shared among all three modules. Frontend
The frontend is a single-page Angular.js app that runs in the browser and interfaces with the API, via both pull (REST) and push (WebSockets) modalities. It is contained in /public/angular
Challenges we ran into
36 hours is too short of a time for any nodejs project and its problems
Accomplishments that we're proud of
What we learned
What's next for HarveyTrack
user: admin pass: chagolchana