Automatically detecting insider threats using user collaboration patterns

1. A computer-implemented method for automatically detecting insider threats using user collaboration patterns, at least a portion of the method being performed by a computing device comprising at least one processor, the method comprising:

identifying collaborative access of one or more network resources in a network between a target user using a target network device and other users using other network devices in the network during multiple prior time periods and during a current time period;

generating prior collaboration graphs for the prior time periods with nodes representing the target user and the other users and edges representing collaborative access of the one or more network resources during the prior time periods between the target user and the other users;

generating an average collaboration graph by combining the prior collaboration graphs, the average collaboration graph including an average number of nodes and an average number of edges from the prior collaboration graphs, with nodes and edges of the average collaboration graph being probabilistically chosen based on their frequency and recency of occurrence;

generating a current collaboration graph for the current time period with nodes representing the target user and the other users and edges representing collaborative access of the one or more network resources during the current time period between the target user and the other users;

generating an anomaly score by comparing the current collaboration graph to the average collaboration graph;

determining that the collaborative access of the one or more network resources during the current time period is anomalous by determining that the anomaly score exceeds a threshold based on the current collaboration graph being larger or smaller in size than, or having a different composition than, the average collaboration graph; and

in response to the anomaly score exceeding the threshold, performing a security action on the target network device.

2. The method of claim 1 , wherein the one or more network resources comprise one or more of an electronic file, online documentation, a network device, and a database.

3. The method of claim 1 , wherein the generating of the average collaboration graph comprises generating a k-step neighborhood network of the prior collaboration graphs.

4. The method of claim 1 , wherein the generating of the anomaly score comprises determining a distance between the current collaboration graph and the average collaboration graph.

5. The method of claim 1 , wherein the performing of the security action on the target network device comprises one or more of blocking the target network device from having access to a website, at least partially blocking the target network device from having access to the network, at least partially blocking the target network device from sending data over the network, and preventing the target network device from accessing email.

6. The method of claim 1 , wherein the determining that the anomaly score exceeds the threshold is based on the current collaboration graph being larger or smaller in size than the average collaboration graph.

7. The method of claim 1 , wherein the determining that the anomaly score exceeds the threshold is based on the current collaboration graph having a different composition than the average collaboration graph.

8. The method of claim 1 , wherein the method is performed in one or more of a User Behavior Analytics (UBA) application and a Security Information and Event Management (SIEM) application.

9. One or more non-transitory computer-readable media comprising one or more computer-readable instructions that, when executed by one or more processors of one or more computing devices, cause the one or more computing devices to perform a method for automatically detecting insider threats using user collaboration patterns, the method comprising:

identifying collaborative access of one or more network resources in a network between a target user using a target network device and other users using other network devices in the network during multiple prior time periods and during a current time period;

generating prior collaboration graphs for the prior time periods with nodes representing the target user and the other users and edges representing collaborative access of the one or more network resources during the prior time periods between the target user and the other users;

generating an average collaboration graph by combining the prior collaboration graphs, the average collaboration graph including an average number of nodes and an average number of edges from the prior collaboration graphs, with nodes and edges of the average collaboration graph being probabilistically chosen based on their frequency and recency of occurrence;

generating a current collaboration graph for the current time period with nodes representing the target user and the other users and edges representing collaborative access of the one or more network resources during the current time period between the target user and the other users;

generating an anomaly score by comparing the current collaboration graph to the average collaboration graph;

determining that the collaborative access of the one or more network resources during the current time period is anomalous by determining that the anomaly score exceeds a threshold based on the current collaboration graph being larger or smaller in size than, or having a different composition than, the average collaboration graph; and

in response to the anomaly score exceeding the threshold, performing a security action on the target network device.

10. The one or more non-transitory computer-readable media of claim 9 , wherein the one or more network resources comprise one or more of an electronic file, online documentation, a network device, and a database.

11. The one or more non-transitory computer-readable media of claim 9 , wherein the generating of the average collaboration graph comprises generating a k-step neighborhood network of the prior collaboration graphs.

12. The one or more non-transitory computer-readable media of claim 9 wherein the generating of the anomaly score comprises determining a distance between the current collaboration graph and the average collaboration graph.

13. The one or more non-transitory computer-readable media of claim 9 , wherein the performing of the security action on the target network device comprises one or more of blocking the target network device from having access to a website, at least partially blocking the target network device from having access to the network, at least partially blocking the target network device from sending data over the network, and preventing the target network device from accessing email.

14. The one or more non-transitory computer-readable media of claim 9 , wherein the determining that the anomaly score exceeds the threshold is based on the current collaboration graph being larger or smaller in size than the average collaboration graph.

15. The one or more non-transitory computer-readable media of claim 9 , wherein the determining that the anomaly score exceeds the threshold is based on the current collaboration graph having a different composition than the average collaboration graph.

16. The one or more non-transitory computer-readable media of claim 9 , wherein the method is performed in one or more of a User Behavior Analytics (UBA) application and a Security Information and Event Management (SIEM) application.