Inspiration
Most of the popular news offer censored content, which is in thier full control. We want to make a platform that will offer most relevant content, without censorship, fresh from the web itself!
What it does
It utilizes social media feeds in order to generate news articles based on recent & popular events and events that are relevant to the user.
How we built it
We built a server which analyzes users preferences from their facebook feed, via unsupervised learning, for the purpose of generating user personally tailored content. While scrapping the twitter data, we use a tool called SynthaxNet, which gives us a complete text semantics, which we use to properly serve twitter feed. Our main goal is to one day build a autonomous news generating service.
Challenges we ran into
Parsing text fetched from Twitter API, and preparing it for the SynthaxNet. Organization and implementation of server side logic, and algorithm design for personally filtered feeds. Building a unique and interactive UX.
Accomplishments that we're proud of
Integrating a lot of API's into our core. Dealt with SynthaxNet to the end. Built an amazing UI. Slowly, but surely we're getting the hand of NLP, and we're closer and closer to building our own AI sentence generator.
What we learned
We learned Natural Language Processing, Twitter API handling, Facebook Graph API, Material Design, RedisDB.
What's next for Citizen Reporter
We'll continue iterating & improving our product, until we get it where it needs to be. Clean news is something people always wanted, and something other than the human mind might finally provide this. We also plan to integrate all the major social media, like Linkedin and Reddit.
Log in or sign up for Devpost to join the conversation.