ML.NET Hackathon Idea - Twitter DM Filter

Hackathon Idea - Twitter DM Filter

Silencer™ a browser plugin, that filters out hateful messages from your twitter DMs.

Your name

Who's submitting this idea? If you already have a team, add all members of the team here

Team name

Silencer™

Brief Description

Twitter is the best website on the world. There are thousands of insightful people, who offer wonderful insight and the most recent news about technology 😊 Twitter is the worst website on the world. There are thousands of bot accounts, real actual fascists, sexists and other insufferable individuals, that destroy your DMs with an endless stream of vitriol (and/or worse).

I'd like to get rid of the second group and keep the first one. Especially in my DMs. That is why I want to build Silencer™ a browser extension, that scans my DMS while I'm in then and hides/flags/beautifies the messages, that would cause me mental anguish otherwise.

Because I like my privacy I don't want to send those DMs to a server I don't trust. I want to do this locally. The plan is to create a browser extension using client side Blazor (or Bolero if I manage to get the Blazor bits, that allow me to use mlnet inference). I saw one working example of a Blazor browser extension once...so I guess it should be possible.

Other

Are you looking for team members?

[ ] Yes
[ ] No
[x] I don't know yet because I have to do some tests first

Would you like to have a mentor assigned to your team?

[ ] Yes
[ ] No
[x] I don't know yet because I have to do some tests first

virtualmlnet / hackathon-2020