Granted Patent B2

US 9,646,155 · App. 14/239,885 · Granted May 9, 2017

Abstract

Claims (47)

1. A method for evaluation of events, implemented by a physical processor executing computer readable instructions, the method comprising:

generating, by the processor, a user-specific reference baseline comprising a set of temporally-ordered sequences of events identified from page-by-page traversal data of the user through a financial system, wherein the reference baseline includes timing statistics for each of the sequence of events;

receiving, by the processor, an event of a sequence of events in a current session;

determining, by the processor, whether the event at least partially matches the reference baseline using an attribute of the event and based on a temporal position of the event within the sequence of events in the current session; and

analyzing, by the processor, a condition of a rule based on the determination of whether the event at least partially matches the reference baseline.

2. The method of claim 1 , wherein the reference baseline further comprises timing data associated with each sequence in the set of temporally-ordered sequences of events, wherein each sequence in the set is a distinctive pattern of historical events associated with the user.

3. The method of claim 1 , wherein determining, by the processor, whether the event at least partially matches the reference baseline comprises:

identifying, by the processor, the user-specific reference baseline among a plurality of reference baselines based on a user identifier associated with the received event.

4. The method of claim 1 , wherein determining, by the processor, whether the event at least partially matches the reference baseline further comprises determining whether the attribute of the received event is within a deviation threshold specified in the condition of the rule.

5. The method of claim 1 , wherein at least one sequence in the set of temporally-ordered sequences of events is flagged, by the processor, as being either normal user behavior or anomalous user behavior.

6. The method of claim 5 , wherein analyzing, by the processor, the condition of the rule comprises:

determining, by the processor, the rule condition is “false” if the event matches the at least one sequence which is flagged as normal user behavior; and

determining, by the processor, the rule condition is “true” if the event matches the at least one sequence which is flagged as anomalous user behavior.

7. The method of claim 1 , wherein determining, by the processor, whether the received event at least partially matches the reference baseline comprises:

tracking, by the processor, a time gap between a time of occurrence of the received event and a time of occurrence of a previously received event of the sequence of events in the current session.

8. The method of claim 1 , wherein each sequence in the set of temporally-ordered sequences of events is associated with timing data generated using a time of occurrence of events in a raw baseline.

9. A system for evaluation of events, the system comprising:

a physical processor implementing machine readable instructions stored on a non-transitory machine-readable storage medium that cause the system to:

store a data list including a plurality of user-specific reference baselines;

generate a first user-specific reference baseline of the plurality of user-specific reference baselines, the first user-specific reference baseline comprising a set of temporally-ordered sequences of events, wherein each reference baseline includes timing statistics for each of the sequence of events;

receive an event of a sequence of events in a current session;

determine whether the event at least partially matches the reference baseline using an attribute of the event and based on a temporal position of the event within the sequence of events in the current session;

analyze a condition of a rule based on the determination of whether the event at least partially matches the reference baseline; and

generate a correlation event.

10. The system of claim 9 , wherein the reference baseline further comprises timing data associated with each sequence in the set of temporally-ordered sequences of events, wherein each sequence in the set is a distinctive pattern of historical events associated with the user.

11. The system of claim 9 , wherein the physical processor implements machine readable instructions that cause the system to determine whether the event at least partially matches the reference baseline by:

identifying the user-specific reference baseline among a plurality of reference baselines based on a user identifier associated with the received event;

selecting an event in a sequence of the reference baseline having the same temporal position as the temporal position of the received event;

comparing the attribute of the received event to a same attribute of the event in the reference baseline, wherein the attribute of the received event is an event type.

12. The system of claim 11 , wherein the physical processor implements machine readable instructions that cause the system to determine whether the event at least partially matches the reference baseline by:

determining whether the attribute of the received event is within a deviation threshold specified in the condition of the rule.

13. The system of claim 9 , wherein at least one sequence in the set of temporally-ordered sequences of events is flagged as being either normal user behavior or anomalous user behavior.

14. A non-transitory computer-readable medium storing a plurality of instructions to control a data processor to evaluate events, the plurality of instructions comprising instructions that cause the data processor to:

generate a user-specific reference baseline, the reference baseline comprising a set of temporally-ordered sequences of transactional events identified from page-by-page traversal data of the user through a financial system, wherein the reference baseline includes timing statistics for each of the sequence of events;

receive a transactional event of a sequence of transactional events in a current session;

determine whether the transactional event at least partially matches the reference baseline using an attribute of the transactional event and based on a temporal position of the transactional event within the sequence of transactional events in the current session;

analyze a condition of a rule based on the determination of whether the transactional event at least partially matches the reference baseline; and

correlate the transactional event with a security-related event.

15. The non-transitory computer-readable medium of claim 14 , wherein the instructions that cause the data processor to determine whether the event at least partially matches the reference baseline comprise instructions that cause the data processor to:

identify the user-specific reference baseline among a plurality of reference baselines based on a user identifier associated with the received event;

select an event in a sequence of the reference baseline having the same temporal position as the temporal position of the received event; and

compare the attribute of the received event to a same attribute of the event in the reference baseline, wherein the attribute of the received event is an event type.

16. The method of claim 1 , further comprising determining, by the processor, an average time gap from a previous event in the sequence of events, wherein the timing statistics includes the determined average time gap.

17. The method of claim 1 , further comprising determining, by the processor, a standard deviation time gap from a previous event in the sequence of events, wherein the timing statistics includes the determined standard deviation time gap.

18. The method of claim 7 , wherein determining, by the processor, whether the received event at least partially matches the reference baseline comprises comparing, by the processor, the time gap to timing data of the reference baseline.

19. The method of claim 1 , wherein determining, by the processor, whether the event at least partially matches the reference baseline comprises selecting, by the processor, an event in a sequence of the reference baseline having the same temporal position as the temporal position of the received event.

20. The method of claim 1 , wherein determining, by the processor, whether the event at least partially matches the reference baseline comprises comparing, by the processor, the attribute of the received event to a same attribute of the event in the reference baseline, wherein the attribute of the received event is an event type.

Assignments (13)