Chaos Labs CEO Shares the Analysis Principles and Methods of Sybil

Chaos Labs' CEO Omer shared the principles and methods of Sybil analysis. His goal is to focus on industrialized Farmers, mainly relying on heuristic methods inspired by funding source analysis. The number of EVM users in Layer Zero is 4.82 million, about 14.5% of whom are listed as Sybil users. Identification methods include detecting fund sources, initial entity identification (users who receive funds from the same source within 24 hours), detecting abnormal funds (providing funds for more than X wallets within 24 hours, Txs from the same source and similar behavior), behavior patterns/similarity vectors for clustering, expanding behavior patterns (transaction volume, number of applications on-chain, multi-chain similar activities and average transaction volume), further strictness in cluster recognition, Sybil classifiers with the same funding source - concentration requirements (Silhouette Score>0.7), Sybil classifier – size requirements (meeting both relative scale and threshold size requirements at the same time).