Inspiration
Social media users have experienced horrendous cases of racial abuse which affected their mental health and wellbeing. Taking this as inspiration, we came up with a solution to tackle this issue on various social media platforms.
What it does
We conceptualized an idea where social media posts containing racial slurs would be flagged and potentially taken down. In addition, it would try and educate the offensive user against racial practices by displaying content which recognizes the struggles of the targeted race and celebrates their achievements.
How we built it
Our AI algorithm is conceptualized as one having 3 layers which identifies the racially abusive posts via direct text matching, color coding and pattern identification to track down a broader range of abusive comments. We coded a very small sample of this algorithm, in which it categorizes the words of a comment into various colors, in order to express the extent of its derogatory nature.
Challenges we ran into
We had issues with time management i.e. we could have built on this project to provide more detail had we managed our time well.
Accomplishments that we're proud of
We figured a 3 layered efficient algorithm which identifies racially offensive comments on social media.
What we learned
For developing our idea, we had to browse through various articles on the basic of AI and machine learning, and this process helped us learn a lot about the same and guided us in the conceptualization of our idea.
What's next for Clasped
We hope to convert this idea into a solution and we envision it to be used in social media platforms to ensure equality and eradicate abuse.
Built With
- ai
- python
Log in or sign up for Devpost to join the conversation.