Method and apparatus for classifying and combining computer attack information
A method and apparatus for classifying and combining computer attack information identifying as malicious events, events in a network that cause organizationally or functionally distant entities to become closer to each other, the method comprising identifying as malicious events, events in a network that cause organizationally or functionally distant entities to become closer to each other.
1. A computer-implemented method of classifying computer attack information performed by a computerized device, comprising:
monitoring a set of entities and events in a computer network;
determining by the computerized device a static distance between two entities, wherein the static distance is defined by an extent 1) that the two entities belong to a same organizational unit or department; and 2) that the two entities share common installed software components; and
determining by the computerized device a dynamic distance between the two entities, wherein the dynamic distance is defined by an extent of 1) a number of common privilege changes on each of the entities 2) a number of common websites accessed by users of the entities; and 3) a number of common newly installed software applications on the entities; and
combining the static distance and the dynamic distance to obtain each of a first entity distance and second entity distance, the first entity distance being computed without an event and the second entity distance being computed with the event;
wherein when the second entity distance is lower than the first entity distance, the event is classified as malicious;
outputting a result of the classification.
2. The method of claim 1 , further comprising associating a rank with the event.
3. The method of claim 2 , further comprising determining at least one cluster of entities with respect to the second entity distance, and associating a cluster rank with each cluster, based on the rank associated with the event.
4. The method of claim 2 , wherein the rank of the event is based on comparing a distance between an entity associated with the event and other entities with and without the event.
5. The method of claim 1 , further comprising associating events in at least one event cluster with malicious activity.
6. The method of claim 5 , further comprising sending an alert associated with at least one event in the at least one event cluster.
7. The method of claim 1 , further comprising discarding the first entity distance or the second entity distance when low value is obtained.
8. The method of claim 1 , further comprising discarding the first entity distance or the second entity distance when high value is obtained.
9. The method of claim 1 , further comprising reducing from the first entity distance or the second entity distance a previous distance value.
10. The method of claim 1 , wherein the entities are selected from the group consisting of: a computer; a process; a thread; software; hardware; a module; a service; an application; a database; a web site; and a user.
11. The method of claim 1 , wherein the event is selected from the group consisting of: changing a computer privilege; installing software; activating software; executing a process; and accessing a web site.
12. The method of claim 1 , further comprising receiving a description of the computer network, the description comprising at least one item selected from the group consisting of: division of computers in the computer network into organizational units, users of each computer; privileges of each user on a computer; software installed on each computer; and web sites accessed by each computer.
13. An apparatus for classifying computer attack information, comprising:
a central processing unit;
an entity distance determination component for selecting at least two entities of a computer network affected by at least one event, determining at least a first entity distance between the at least two entities without the at least one event, and for determining at least a second entity distance between the at least two entities with the at least one event;
wherein determining the at least one first entity distance and the at least one second entity distance comprises:
determining by a computer a static distance between the at least two entities, wherein the static distance is defined by an extent 1) that the at least two entities belong to a same organizational unit or department; and 2) that the at least two entities share common installed software components; and
determining by the computer a dynamic distance between the at least two entities, wherein the dynamic distance is defined by an extent of 1) a number of common privilege changes on each of the at least two entities 2) a number of common websites accessed by users of the at least two entities; and 3) a number of common newly installed software applications on the at least two entities; and
combining the static distance and the dynamic distance to obtain each of the first entity distance and the second entity distance;
wherein when the at least one second entity distance is lower than the at least one first entity distance, the at least one event is classified as malicious;
wherein the entity distance determination component is implemented by the central processing unit.
14. The apparatus of claim 13 , further comprising an event rank determination component for determining a rank associated with the at least one event, and a cluster rank determination component for determining a rank associated with a cluster; wherein each of the event rank determination component and the cluster rank determination component is implemented by the central processing unit.
15. The apparatus of claim 14 , wherein the event rank determination component and the cluster rank determination component determines the rank associated with the event based on comparing the first entity distance and the second entity distance.
16. The apparatus of claim 13 , further comprising a cluster determination component for determining at least one event cluster; wherein the cluster determination component is implemented by the central processing unit.
17. The apparatus of claim 16 , further comprising an alert generation component for generating an alert related to at least one of the at least one event cluster; wherein the alert generation component is implemented by the central processing unit.
18. The apparatus of claim 13 , wherein the entity distance determination component is configured to determine the at least first entity distance and the at least second entity distance for each entity pair of organizationally or functionally distant entities of a multiplicity of entity pairs of the computer network.
19. A computer program product for classifying computer attack information, comprising:
a non-transitory computer readable medium;
a program instruction for determining an entity distance between at least two entities, wherein the program instruction is stored on said non-transitory computer readable medium;
wherein determining the entity distance comprises:
determining by the computer program product a static distance between the at least two entities, wherein the static distance is defined by an extent 1) that the at least two entities belong to a same organizational unit or department; and 2) that the at least two entities share common installed software components; and
determining by the computer program product a dynamic distance between the at least two entities, wherein the dynamic distance is defined by an extent of 1) a number of common privilege changes on each of the at least two entities 2) a number of common websites accessed by users of the at least two entities; and 3) a number of common newly installed software applications on the at least two entities; and
combining the static distance and the dynamic distance to obtain each of a first distance and a second distance, the first distance being computed without an event and the second distance being computed with the event;
wherein when the second distance is lower than the first distance, the event is classified as malicious.