Granted Patent B2

US 12,244,620 · App. 17/815,963 · Granted Mar 4, 2025

Inventors: Rama Krishnam Raju Rudraraju (Telangana, IN); Om Purushotham Akarapu (Telangana, IN)

Assignee: Bank of America Corporation

H04L63/1425

Abstract

Claims (95)

1. A system for implementing anomaly detection, comprising:

a memory configured to store first user activities associated with an avatar within a first virtual environment, wherein:

the avatar is associated with a user; and

the first user activities comprise one or more first interactions between the avatar and other entities in the first virtual environment; and

a processor operably coupled with the memory, and configured to:

access the first user activities;

extract a first set of features from the first user activities, wherein the first set of features provides information about at least the one or more first interactions;

for each feature of the first set of features, determine a first deviation range that indicates a deviation between a selected feature associated with the user and the selected feature associated with one or more other users over a certain period;

assign the selected feature to a second set of features when the first deviation range is more than a threshold deviation;

determine a priority weight for each feature included in the second set of features;

apply the priority weight to the first deviation range of each feature included in the second set of features to determine a weighted deviation range of each feature;

determine a combined deviation range by combining the weighted deviation range of each feature included in the second set of features; and

determine a confidence score associated with the user based at least in part upon the combined deviation range, wherein the confidence score indicates whether the user is associated with an anomaly, such that:

if the confidence score is more than a threshold percentage, the user is not associated with an anomaly; and

if the confidence score is less than the threshold percentage, the user is associated with the anomaly.

2. The system of claim 1 , wherein in response to determining that the first deviation range is less than the threshold deviation, the confidence score is not updated based on the selected feature.

3. The system of claim 1 , wherein the processor is further configured to:

access second user activities associated with the avatar within a second virtual environment, wherein the second user activities comprise one or more second interactions between the avatar and other entities in the second virtual environment;

extract a second set of features from the second user activities, wherein the second set of features provides information about at least the one or more second interactions;

for a second feature from among the second set of features, determine a second deviation range that indicates a deviation between a maximum value and a minimum value of the second feature over the certain period;

determine whether the second deviation range is more than the threshold deviation; and

in response to determining that the second deviation range is more than the threshold deviation, update the confidence score based at least in part upon the second feature.

4. The system of claim 3 , wherein updating the confidence score based at least in part upon the second feature comprises:

increasing the confidence score proportional to the first deviation range; or

decreasing the confidence score proportional to the first deviation range.

5. The system of claim 1 , wherein the processor is further configured to:

determine that the user requests to perform an interaction with an entity in the first virtual environment;

determine whether the confidence score is more than a threshold score; and

in response to determining that the confidence score is more than the threshold score, authorize the user to perform the interaction with the entity.

6. The system of claim 5 , wherein the processor is further configured to, in response to determining that the confidence score is less than the threshold score, prevent the user to perform the interaction with the entity.

7. The system of claim 1 , wherein the first set of features comprises at least one of:

a frequency of historical interactions with the other entities; and

a number of historical interactions with the other entities.

8. A method for implementing anomaly detection, comprising:

accessing first user activities associated with an avatar within a first virtual environment, wherein:

the avatar is associated with a user; and

the first user activities comprise one or more first interactions between the avatar and other entities in the first virtual environment;

extracting a first set of features from the first user activities, wherein the first set of features provides information about at least the one or more first interactions;

for each feature of the first set of features, determining a first deviation range that indicates a deviation between a selected feature associated with the user and the selected feature associated with one or more other users over a certain period;

assigning the selected feature to a second set of features when the first deviation range is more than a threshold deviation;

determining a priority weight for each feature included in the second set of features;

applying the priority weight to the first deviation range of each feature included in the second set of features to determine a weighted deviation range of each feature;

determining a combined deviation range by combining the weighted deviation range of each feature included in the second set of features; and

determining a confidence score associated with the user based at least in part upon the combined deviation range, wherein the confidence score indicates whether the user is associated with an anomaly, such that:

if the confidence score is more than a threshold percentage, the user is not associated with an anomaly; and

if the confidence score is less than the threshold percentage, the user is associated with the anomaly.

9. The method of claim 8 , wherein in response to determining that the first deviation range is less than the threshold deviation, the confidence score is not updated based on the selected feature.

10. The method of claim 8 , further comprising:

accessing second user activities associated with the avatar within a second virtual environment, wherein the second user activities comprise one or more second interactions between the avatar and other entities in the second virtual environment;

extracting a second set of features from the second user activities, wherein the second set of features provides information about at least the one or more second interactions;

for a second feature from among the second set of features, determining a second deviation range that indicates a deviation between a maximum value and a minimum value of the second feature over the certain period;

determining whether the second deviation range is more than the threshold deviation; and

in response to determining that the second deviation range is more than the threshold deviation, updating the confidence score based at least in part upon the second feature.

11. The method of claim 10 , wherein updating the confidence score based at least in part upon the second first feature comprises:

increasing the confidence score proportional to the deviation range; or

decreasing the confidence score proportional to the first deviation range.

12. The method of claim 8 , further comprising:

determining that the user requests to perform an interaction with an entity in the first virtual environment;

determining whether the confidence score is more than a threshold score; and

in response to determining that the confidence score is more than the threshold score, authorizing the user to perform the interaction with the entity.

13. The method of claim 12 , further comprising, in response to determining that the confidence score is less than the threshold score, preventing the user to perform the interaction with the entity.

14. The method of claim 8 , wherein the first set of features comprises at least one of:

a frequency of historical interactions with the other entities; and

a number of historical interactions with the other entities.

15. A non-transitory computer-readable medium that stores instructions, wherein when the instructions are executed by one or more processors, cause the one or more processors to:

access first user activities associated with an avatar within a first virtual environment, wherein:

the avatar is associated with a user; and

the first user activities comprise one or more first interactions between the avatar and other entities in the first virtual environment; and

extract a first set of features from the first user activities, wherein the first set of features provides information about at least the one or more first interactions;

for each feature of the first set of features, determine a first deviation range that indicates a deviation between a selected feature associated with the user and the selected feature associated with one or more other users over a certain period;

assign the selected feature to a second set of features when the first deviation range is more than a threshold deviation;

determine a priority weight for each feature included in the second set of features;

apply the priority weight to the first deviation range of each feature included in the second set of features to determine a weighted deviation range of each feature;

determine a combined deviation range by combining the weighted deviation range of each feature included in the second set of features; and

determine a confidence score associated with the user based at least in part upon the combined deviation range, wherein the confidence score indicates whether the user is associated with an anomaly, such that:

if the confidence score is more than a threshold percentage, the user is not associated with an anomaly; and

if the confidence score is less than the threshold percentage, the user is associated with the anomaly.

16. The non-transitory computer-readable medium of claim 15 , wherein in response to determining that the first deviation range is less than the threshold deviation, the confidence score is not updated based on the selected feature.

17. The non-transitory computer-readable medium of claim 15 , wherein the instructions when executed by the one or more processors, further cause the one or more processors to:

access second user activities associated with the avatar within a second virtual environment, wherein the second user activities comprise one or more second interactions between the avatar and other entities in the second virtual environment;

extract a second set of features from the second user activities, wherein the second set of features provides information about at least the one or more second interactions;

for a second feature from among the second set of features, determine a second deviation range that indicates a deviation between a maximum value and a minimum value of the second feature over the certain period;

determine whether the second deviation range is more than the threshold deviation; and

in response to determining that the second deviation range is more than the threshold deviation, update the confidence score based at least in part upon the second feature.

18. The non-transitory computer-readable medium of claim 17 , wherein updating the confidence score based at least in part upon the second feature comprises:

increasing the confidence score proportional to the first deviation range; or

decreasing the confidence score proportional to the first deviation range.

19. The non-transitory computer-readable medium of claim 15 , wherein the instructions when executed by the one or more processors, further cause the one or more processors to:

determine that the user requests to perform an interaction with an entity in the first virtual environment;

determine whether the confidence score is more than a threshold score; and

in response to determining that the confidence score is more than the threshold score, authorize the user to perform the interaction with the entity.

20. The non-transitory computer-readable medium of claim 15 , wherein determining the first deviation range comprises:

determining a first value of the selected feature associated with the user;

determining a second value of the selected feature associated with the one or more other users; and

determining a difference between the first value with the second value, wherein the difference between the first value and the second value is the first deviation range.

Assignments (1)

Continuity (1)

References Cited (44)