Usage of TrustScan's Sybil Score
Last updated
Last updated
The queryRiskSummaryScore API accepts an EOA address and Chain ID as input. It responds with an overall Sybil score and sub-scores for the specified address on that blockchain. This document explains the API's output Sybil score and provides data-driven guidance on how to interpret and use the score.
The output Sybil score has the value ranged from 0 to 100, where the higher the score, the more Sybil risk of the addresses. Besides, there are two negative integers -2 and -1 of the score to indicate two corner cases. We explain the value in the following table.
To analyze the distribution of risk scores, we conducted a simulation encompassing all active addresses that had at least one on-chain interaction between July 6, 2023 and August 6, 2023. During this period, a total of 5,140,742 unique addresses were active and scored. The full distribution of risk scores is as follows:
If all 5 million active addresses applied for TrustScan's Non-Sybil Stamp, our analysis indicates that approximately 494,226(=469903+15216+9107) addresses or 9.62% would qualify for issuance.
sybilRiskScore
sybilRiskLevel
Explanation
Recommendation
(Non-Sybil Account Verification)
-2
Unknown
Out of scope for TrustScan's Sybil analysis due to insufficient on-chain activity.(With fewer than 3 transactions, TrustScan does not have enough behavioral data points to make a reliable risk assessment.)
False
-1
No Risk
High Value addresses,No apparent Sybil risk detected by TrustScan(Some hub addresses and their connected addresses exhibit high volumes of activity and interactions, leading them to be accessed as high value accounts.)
True
0 - 60
No Risk
No signs of Sybil risk detected
True
60 -80
Low
LOW Sybil risk detected
False
80 - 90
Medium
MEDIUM Sybil risk detected
False
90 -100
High
HIGH Sybil risk detected
False
sybil_score_bin
addr_cnt
ratio
Recommendation (Non-Sybil Account Verification)
-2
4,515,395
87.84%
False
-1
469,903
9.14%
True
[0, 40]
15,216
0.30%
True
(40, 60]
9,107
0.18%
True
(60, 80]
28,591
0.56%
False
(80, 90]
10,644
0.21%
False
(90, 100]
91,886
1.79%
False