Content
@
0 reply
20 recasts
20 reactions
C O M P Ξ Z
@compez.eth
What if we implement a formula like this? I’ve put a lot of thought into designing a formula that evaluates multiple sources and then reaches a balanced conclusion. The idea is that if one source is inaccurate or biased, the others should help correct the final outcome. This approach considers three key inputs: 1️⃣ WC Spam Label – Detects the likelihood of spammy behavior. 2️⃣ OpenRank – Measures engagement and following credibility. 3️⃣ Neynar Score – Evaluates overall user quality based on interactions. By combining these sources mathematically, we get an objective and transparent way to classify users. The output so far looks promising—it provides clear differentiation between trusted users, suspicious accounts, and spammers. What do you think? Would this model make trust scoring more fair and convincing for everyone?
10 replies
7 recasts
32 reactions
mvr 🐹
@mvr
I guess you could add the social score from Moxie as well (without the boost from creator token locks) I think that weights heavy on active followers
1 reply
0 recast
5 reactions
C O M P Ξ Z
@compez.eth
I couldn't find a formula for it! I was thinking about that, but airstack determines rankings based on token dependency.
2 replies
0 recast
1 reaction
Pichi 🟪🍖🐹🎩 🍡🌸
@pichi
@ipeciura.eth tagging you in from FarScore. Compez is trying to combine all the public data sets into trust scores. Not sure if Organic FarScore fits in here but would love your thoughts.
1 reply
0 recast
2 reactions
mvr 🐹
@mvr
The original one which they still have in their responses is without token dependency
0 reply
0 recast
1 reaction
