Nostraterrestrial on Nostr: Anyone care to create a little WoT tool? #asknostr Let’s imagine I follow n=5 ...
Anyone care to create a little WoT tool?
#asknostr
Let’s imagine I follow n=5 people:
A, B, C, D and E
Out of the people I follow:
A is followed by (A, B, C, D, E)
B is followed by (A, B, C)
C is followed by (C, D, E)
D is followed by (D)
E is followed by (E)
Stage 1:
Score each person I follow (n) by how many of the people I follow (my friends) also follow them (m).
score1 = m/n
A: 1.0
B: 0.6
C: 0.6
D: 0.2
E: 0.2
Stage 2:
Score each of my friends by the score1s of my friends that also follow them.
score2 = (sum of score1s of m)/n
A: (1.0 + 0.6 + 0.6 + 0.2 + 0.2)/5 = 0.52
B: (1.0 + 0.6 + 0.6)/5 = 0.44
C: (0.6 + 0.2 + 0.2)/5 = 0.2
D: 0.2/5 = 0.04
E: 0.2/5 = 0.04
Stage 3:
There are three more people (X, Y, Z) in the network. I don’t follow any of them yet.
X is followed by (A, B, E)
Y is followed by (B, C, D, E)
Z is followed by none of them
Score each person that I don’t follow, but is followed by one of my friends by the score2s of my friends that follow them.
X: (0.52 + 0.44 + 0.04)/5 = 0.2
Y: (0.44 + 0.2 + 0.04 + 0.04)/5 = 0.144
Z: 0/5 = 0
Then show me a list of these friends of friends, ranked by their score and a feed of their posts that I can filter by one of these criteria:
- minimum rank (e.g. 0.03 or more)
- top percentage by rank (e.g. top 5%)
- top number by rank (e.g. top 50 people)
Published at
2024-02-17 13:24:24Event JSON
{
"id": "8cbff8b969383ff8fe0ff1d4ab63e4cacd034f6ad4c106c2ddb9f86e45368ec5",
"pubkey": "546a3f8d791a84a4879dc30d4ae9d4ed0aef3d0e1492be649aeda2337be80648",
"created_at": 1708176264,
"kind": 1,
"tags": [
[
"t",
"asknostr"
]
],
"content": "Anyone care to create a little WoT tool?\n#asknostr\n\nLet’s imagine I follow n=5 people:\nA, B, C, D and E\n\nOut of the people I follow:\nA is followed by (A, B, C, D, E)\nB is followed by (A, B, C)\nC is followed by (C, D, E)\nD is followed by (D)\nE is followed by (E)\n\n\nStage 1:\nScore each person I follow (n) by how many of the people I follow (my friends) also follow them (m).\n\nscore1 = m/n\n\nA: 1.0\nB: 0.6\nC: 0.6\nD: 0.2\nE: 0.2\n\n\nStage 2:\nScore each of my friends by the score1s of my friends that also follow them.\n\nscore2 = (sum of score1s of m)/n\n\nA: (1.0 + 0.6 + 0.6 + 0.2 + 0.2)/5 = 0.52\nB: (1.0 + 0.6 + 0.6)/5 = 0.44\nC: (0.6 + 0.2 + 0.2)/5 = 0.2\nD: 0.2/5 = 0.04\nE: 0.2/5 = 0.04\n\n\nStage 3:\nThere are three more people (X, Y, Z) in the network. I don’t follow any of them yet.\nX is followed by (A, B, E)\nY is followed by (B, C, D, E)\nZ is followed by none of them\n\nScore each person that I don’t follow, but is followed by one of my friends by the score2s of my friends that follow them.\n\nX: (0.52 + 0.44 + 0.04)/5 = 0.2\nY: (0.44 + 0.2 + 0.04 + 0.04)/5 = 0.144\nZ: 0/5 = 0\n\n\nThen show me a list of these friends of friends, ranked by their score and a feed of their posts that I can filter by one of these criteria:\n- minimum rank (e.g. 0.03 or more)\n- top percentage by rank (e.g. top 5%)\n- top number by rank (e.g. top 50 people)",
"sig": "39aaf4d440e044c3ee3b84c9f2f5d555f5df7ed3612d53ea2984b90200f6c3c382779e2e9e1dc14ce4edc5309483ced0ccd985cb4866680f598b94ede0b4b10a"
}