[thinkingemote]

There is a "mute" setting built into Twitter which you can populate with words and phrases and which will remove tweets containing those.

It's very powerful and easy to use but it seems that many early adopters are unaware of this simple setting! It's not fuzzy nor able to handle regex so a clever AI solution like this would fill that gap.

(I found that simply adding "trump" in 2016 improved my experience)

[Navkun]

Yeah I used the mute feature extensively (I must have 100+ muted words) but I still saw things that I didn't want to see, quite often, like ramblings of politics and outrage etc...

Now I have muted words + mindfirewall, I never see anything I don't want to.

[thinkingemote]

Yes it's a crude and blunt tool, your solution add good flexibility to it, bravo! I wonder if we will see more tools like this that uses ... AI discernment? ... to serve what the users desire.