I'm excited about the new @pixelfed Autospam Advanced Detection, but I should warn that your results may vary based on training data.

I did my best to make this as easy as possible, but it really comes down to training models.

That's why I added the ability to import or export training data, though it's currently limited to positive/spam tokens for privacy reasons.

I look forward to shipping an importable community model and improving spam detection overall.

Fuck spam. ๐Ÿ˜Ž

2
Share
Share on Mastodon
Share on Twitter
Share on Facebook
Share on Linkedin
Meow :verified:

@dansup @pixelfed I feel the same way about spam. I wish there was a way that we could actively crash their machines.

0
1y
Patrick P.

@dansup @pixelfed With some chance crypto scams will help soon to train this IA. A good point for the Fediverse popularity. ๐Ÿ˜

0
1y
Replies