Systems and methods for detecting suspicious users on networks
The disclosed computer-implemented method for detecting suspicious users on networks may include (1) identifying a first cluster of users based on at least one static attribute of the users; (2) identifying a second cluster of users based on both the at least one static attribute of the users and at least one dynamic attribute of the users, where a respective dynamic attribute weight applied to the at least one dynamic attribute when defining the second clusters is based on network monitoring telemetry; (3) comparing the first cluster with the second cluster to identify an outlying user; and (4) designating the outlying user as suspicious.
1. A computer-implemented method for detecting suspicious users on networks, at least a portion of the method being performed by a computing device comprising at least one processor, the method comprising:
calculating connection weights between users by adding respective static attribute weights of the users when the users have the same static attribute weights;
identifying a first cluster of users based on the calculated connection weights;
identifying a second cluster of users based on both at least one static attribute of the users and at least one dynamic attribute of the users, wherein the identifying the second cluster of users comprises:
adjusting the calculated connection weight between a set of users to create an adjusted connection weight by:
multiplying the calculated connection weight between the set of users by a predetermined factor when a respective dynamic attribute weight of the set of users is within a predetermined range; and
dividing the calculated connection weight between the set of users by the predetermined factor when the respective dynamic attribute weight of the set of users is outside of the predetermined range; and
identifying the second cluster of users based on the adjusted connection weight;
comparing the first cluster with the second cluster to identify an outlying user; and
designating the outlying user as suspicious.
2. The computer-implemented method of claim 1 , wherein the comparing further comprises comparing distances between locations of each user in the first cluster and the second cluster to identify the outlying user.
3. The computer-implemented method of claim 1 , wherein the at least one static attribute is based on a user profile.
4. The computer-implemented method of claim 3 , wherein the user profile comprises at least one of:
a job title;
a job grade;
a job location;
a job department;
a country; and
a primary language.
5. The computer-implemented method of claim 1 , wherein the at least one dynamic attribute is based on user behavior.
6. The computer-implemented method of claim 5 , wherein the user behavior comprises at least one of:
a number of files accessed;
a number of file permissions;
a number of file reads;
a number of file writes;
a number of file creates;
a number of file deletes;
a number of file permission changes;
a number of file renames;
a number of files to which the user has access;
a number of internet protocol addresses that the user accesses; and
a number of files that the user owns.
7. The computer-implemented method of claim 1 , further comprising performing at least one security action in response to designating the outlying user as suspicious.
8. A system for detecting suspicious users on networks, the system comprising:
an identifying module, stored in memory, that:
calculates connection weights between users by adding respective static attribute weights of the users when the users have the same static attribute weights;
identifies a first cluster of users based on the calculated connection weights; and
identifies a second cluster of users based on both the at least one static attribute of the users and at least one dynamic attribute of the users, wherein the identifying the second cluster of users comprises:
adjusting the calculated connection weight between a set of users to create an adjusted connection weight by:
multiplying the calculated connection weight between the set of users by a predetermined factor when a respective dynamic attribute weight of the set of users is within a predetermined range; and
dividing the calculated connection weight between the set of users by the predetermined factor when the respective dynamic attribute weight of the set of users is outside of the predetermined range; and
identifying the second cluster of users based on the adjusted connection weight;
a comparing module, stored in memory, that compares the first cluster with the second cluster to identify an outlying user;
a designating module, stored in memory, that designates the outlying user as suspicious; and
at least one physical processor that executes the identifying module, the comparing module, and the designating module.
9. The system of claim 8 , wherein the comparing further comprises comparing distances between locations of each user in the first cluster and the second cluster to identify the outlying user.
10. The system of claim 8 , wherein the at least one static attribute is based on a user profile.
11. The system of claim 10 , wherein the user profile comprises at least one of:
a job title;
a job grade;
a job location;
a job department;
a country; and
a primary language.
12. The system of claim 8 , wherein the at least one dynamic attribute is based on user behavior.
13. The system of claim 12 , wherein the user behavior comprises at least one of:
a number of files accessed;
a number of file permissions;
a number of file reads;
a number of file writes;
a number of file creates;
a number of file deletes;
a number of file permission changes;
a number of file renames;
a number of files to which the user has access;
a number of internet protocol addresses that the user accesses; and
a number of files that the user owns.
14. The system of claim 8 , further comprising a security module, stored in memory, that performs at least one security action in response to designating the outlying user as suspicious.
15. A non-transitory computer-readable medium comprising one or more computer-executable instructions that, when executed by at least one processor of a computing device, cause the computing device to:
calculate connection weights between users by adding respective static attribute weights of the users when the users have the same static attribute weights;
identify a first cluster of users based on the calculated connection weights;
identify a second cluster of users based on both at least one static attribute of the users and at least one dynamic attribute of the users, wherein the identifying the second cluster of users comprises:
adjusting the calculated connection weight between a set of users to create an adjusted connection weight by:
multiplying the calculated connection weight between the set of users by a predetermined factor when a respective dynamic attribute weight of the set of users is within a predetermined range; and
dividing the calculated connection weight between the set of users by the predetermined factor when the respective dynamic attribute weight of the set of users is outside of the predetermined range; and
identifying the second cluster of users based on the adjusted connection weight;
compare the first cluster with the second cluster to identify an outlying user; and
designate the outlying user as suspicious.
16. The non-transitory computer-readable medium of claim 15 , wherein the computer-executable instructions that cause the computing device to compare the first cluster with the second cluster further comprise computer-executable instructions that cause the computing device to compare distances between locations of each user in the first cluster and the second cluster to identify the outlying user.
17. The non-transitory computer-readable medium of claim 15 , wherein the at least one static attribute is based on a user profile.
18. The non-transitory computer-readable medium of claim 15 , further comprising computer-executable instructions that cause the computing device to perform at least one security action in response to designating the outlying user as suspicious.
19. The non-transitory computer-readable medium of claim 17 , wherein the user profile comprises at least one of:
a job title;
a job grade;
a job location;
a job department;
a country; and
a primary language.
20. The non-transitory computer-readable medium of claim 15 , further comprising computer-executable instructions that cause the computing device to perform at least one security action in response to designating the outlying user as suspicious.