System and method for machine learning based security incidents detection and classification in a blockchain ecosystem

A system for real-time detection of illicit activity in a blockchain environment, comprises:

• a processing circuitry;
• and
• a memory, the memory containing instructions that, when executed by the processing circuitry, configure the system to:
• collect from one or more electronic sources on-chain data of a plurality of blockchain transactions occurred between a plurality of blockchain addresses;
• collect off-chain data of a plurality of entities, wherein each entity of the plurality of entities is associated with at least one blockchain transaction of the plurality of blockchain transactions;
• generate a cross-chain graph representation based on at least the collected on-chain data of the plurality of blockchain transactions and the off-chain data of the plurality of entities, wherein the cross-chain graph representation includes a plurality of nodes representing blockchain addresses and a plurality of edges representing connections between at least a portion of the plurality of nodes;
• apply to the cross-chain graph representation a machine learning model that is adapted to detect suspicious anomalies in the cross-chain graph representation;
• and
• when one or more suspicious anomalies have been detected above a dynamic model-based threshold value in the cross-chain graph representation, instigate, in real-time, the taking of an action regarding detection of a security incident.