CPC H04L 63/1433 (2013.01) [H04L 63/101 (2013.01); H04L 63/104 (2013.01); H04L 63/1425 (2013.01)] | 20 Claims |
1. A computer-implemented method for assessing risk to a system, the method comprising:
detecting a plurality of events associated with the system, wherein each event is associated with at least one entity of a plurality of entities and at least one attack type of a plurality of attack types;
for each event of the plurality of events, determining a plurality of risk scores, wherein each risk score of the plurality of risk scores corresponds to a separate attack type of the plurality of attack types and the plurality of risk scores for each event are encoded in a first multidimensional (MD) data object, wherein the first MD data object includes a three dimensional tensor where risk scores of the plurality of risk scores are encoded based on an event of the plurality of events, an attack type of the plurality of attack types, and an entity of the plurality of entities such that a first component of the first MD data object includes a vector encoding a subset of the risk scores of the plurality of risk scores corresponding to attack types of the plurality of attack types;
for each entity of the plurality of entities, determining an entity risk score based on the first MD data object by at least summing risk scores included in the first MD data object for a particular entity of the plurality of entities to generate a second MD data object of a lower dimension than the first MD data object;
identifying at least one high-risk entity of the plurality of entities based on the entity risk score included in the second MD data object and at least one risk threshold; and
providing a risk report that includes an indication of the at least one high-risk entity by at least providing multi-dimensional information associated with the at least one high-risk entity, where the multi-dimensional information includes information associated with the plurality of attack types, the plurality of events, and the plurality of entities.
|