Content
@
0 reply
0 recast
0 reaction
Colin
@colin
I think this is a very interesting problem space. It’s similar to some work I was doing at Google: labeling & predicting abusive behavior for millions of SMS messages across 1b phone numbers a month. We would: - manually label millions of spam reports - create heuristic rules & train ML models using labeled data - deploy heuristics & models to the device and on the server (bc different access to signals on each) - rinse & repeat Heuristics got us surprisingly far. The vast amount of accessible onchain and FC data (metadata, behavior, content) should make it much easier to build decent classifiers.
5 replies
2 recasts
30 reactions
Yassine Landa
@yassinelanda.eth
I didn't know this! I think now adversarial bots and spammers are much more sophisticated speaking to people at Meta & Google. Not seeing this on fc a lot yet, but heuristic are okayish. Agree we got to decent classifiers really fast and there is more context to tap into!
0 reply
0 recast
0 reaction
Ape/rture
@aperture
Using heuristics and top down thinking on the data makes the classification process easier because the results just make more sense. The ML model would just be a bonus where there might be some patterns related to those heuristics being picked up. We can't ask the MLs to understand what humans think is valuable in this data (yet) without any guidance. It seems sometimes devs in crypto forget the classic data science cycle: ad hoc research/reports > automate that > then decide if you have do dig deeper with extensive tech like MLs.
0 reply
0 recast
0 reaction
Samir 🎩
@0xsamir
Any specific number of casts to begin labeling and classifying needed? Like we need to classify manually like 10k casts is that good? Or you do it in different way?
0 reply
0 recast
0 reaction
Dvyne🎩
@dummie.eth
Was it specific words used that you were able to highlight these SMS as spam Cause the algo seems to be weird when it comes to spamming on farcaster
0 reply
0 recast
0 reaction
koisose.lol
@koisose
https://warpcast.com/yassinelanda.eth/0x6a9ee471 i think @yassinelanda.eth nailed it just need to maybe add more label like moxie jail, degen farmer and whatnot
1 reply
0 recast
1 reaction
