Our app detects hateful/harmful messages by understanding the sentences and then alerts the senders.
Millions of people are harassed and bullied online, as more people come online the issue is probably going to get worse. Social media sites are not doing enough to protect users. As developers, we need to be consciously creating software which enables a safer environment for everyone online. This involves filtering and banning hateful content, which can be a hard job. So this is an attempt to stop bullying at the program level by asking people to reconsider their messages before they send if the program detects it might be harmful.
What it does
This is an attempt to devise a simple anti-harassment/bullying framework which can be used across different social apps that people build.
How I built it
I used ReactJS for the frontend, NodeJS on the backend, along with libraries such as React Router, Socket.io, MomentJS
Challenges I ran into
Trying to find ways to create multiple connections in chat rooms. Socket.io helped abstract away a lot of the complexities involved with the socket protocol.
Accomplishments that I'm proud of
Creating an application from scratch which acts as a proof of concept.
What I learned
Working with the Microsoft Azure AI platform, and integrating a react/node app with some of its services.
What's next for No Cyber Bullies
Code release, open-source the underlying code which can be integrated into other apps by developers. A tutorial to get people started with working with the Azure services along with the code.