The project is essentially a proof of concept for chat monitoring systems which can be deployed by the authoritarian bodies to prevent spreading of hate speech and avoid possible violent outcomes.The application checks for hate words simultaneously whenthe user is typing and if it does not find any hate speech the message will not be stored anywhere hence maintaining privacy.This type of system can be applied in various chat applications like whatsapp and telegram with small modifications.
SecuroChat: This is the app where users can communicate with each other through textual messages which will be checked at real-time when user is typing and won't allow to send hate/racist messages.SecuroChat is made to simulate a chat application which can be a medium to send inappropriate messages.If a user tries to send a hate messages,their details like physical cordinates(Location) along with the copy of the message.This can be used to investigate the source of hate information and catch hold of people behind it.
AdminHub: This app is used to set keywords that will be checked,both apps are connected to the same real-time database and throughthis app the admin can also send warning mail to users.
The challenges faced were:
As there is no chatting application with a free open API for message reading we had to develop our own app to prove our idea.
Deploying this system and simultaneously not voilating orivacy was the biggest challenge and we solved it by storing only the user's messages that contain hate speech.
Developing 2 apps in such a short time was a big challenge.
Dealing with two apps connected to the same real-time database was difficult and we can improve this process further.
Keeping consistency in the UI of both apps required quite a bit of efforts.
Technologies used
Discussion