Granted Patent B1

US 10,278,074 · App. 14/920,262 · Granted Apr 30, 2019

Abstract

Claims (63)

1. A computer-implemented method for categorizing mobile devices as rooted, at least a portion of the method being performed by a computing device comprising at least one processor, the method comprising:

gathering a set of metadata describing a plurality of rooted mobile devices that have been modified to allow a user to alter protected systems and an additional set of metadata describing a plurality of unrooted mobile devices that have not been modified to allow the user to alter the protected systems;

comparing the set of metadata with the additional set of metadata by processing the metadata and the additional set of metadata through a series of classifiers comprising a machine-learning algorithm and a linear regression algorithm to determine at least one feature that differentiates the rooted mobile devices from the unrooted mobile devices, wherein the at least one feature comprises an application category including one or more malware application types and one or more applications used for potentially malicious purposes;

determining whether the at least one feature is present in metadata that describes an uncategorized mobile device;

categorizing the uncategorized mobile device as a rooted mobile device based on a total weighted value of the at least one feature present in the metadata exceeding a predetermined threshold; and

protecting a secure network against the one or more malware application types and the one or more applications used for potentially malicious purposes potentially present on the rooted device by disallowing the rooted mobile device from connecting to the secure network.

2. The computer-implemented method of claim 1 , wherein the secure network is governed by a security policy that prohibits rooted devices from connecting to the secure network.

3. The computer-implemented method of claim 2 , further comprising:

categorizing an additional mobile device as an unrooted mobile device based on the absence of the at least one feature in metadata that describes the additional mobile device; and

allowing the unrooted mobile device to connect to the secure network.

4. The computer-implemented method of claim 1 , wherein determining whether the at least one feature is present in the metadata that describes the uncategorized mobile device comprises determining, on the uncategorized mobile device, whether the at least one feature is present.

5. The computer-implemented method of claim 1 , wherein determining whether the at least one feature is present in the metadata that describes the uncategorized mobile device comprises determining, at a server, whether the at least one feature is present.

6. The computer-implemented method of claim 1 , wherein the one or more applications used for potentially malicious purposes are installed on the uncategorized mobile device.

7. The computer-implemented method of claim 1 , wherein the at least one feature comprises a pattern of application installation behavior on the uncategorized mobile device that comprises at least one of:

a number of times applications have been installed on the uncategorized mobile device;

a number of times applications have been uninstalled on the uncategorized mobile device;

a number of times applications have been updated on the uncategorized mobile device;

an average time between two consecutive installations or uninstallations of applications on the uncategorized mobile device.

8. The computer-implemented method of claim 1 , wherein the at least one feature comprises a pattern of power consumption behavior of the uncategorized mobile device that comprises at least one of:

a total power consumption by the uncategorized mobile device over a predetermined time period;

an average power consumption per application installed on the uncategorized mobile device;

an average sensor power consumption on the uncategorized mobile device;

an average battery level on the uncategorized mobile device.

9. The computer-implemented method of claim 1 , wherein the at least one feature comprises a pattern of communication behavior of the uncategorized mobile device that comprises at least one of:

a total amount of data sent by the uncategorized mobile device via a network;

an average amount of data sent per day by the uncategorized mobile device via a network;

a total amount of data received by the uncategorized mobile device via a network;

an average amount of data received per day by the uncategorized mobile device via a network;

a total number of Internet Protocol addresses visited by applications installed on the uncategorized mobile device;

a total number of ports connected to by applications installed on the uncategorized mobile device.

10. The computer-implemented method of claim 1 , wherein categorizing the uncategorized mobile device as a rooted mobile devices comprises at least one of:

displaying a notification on the rooted mobile device indicating that the rooted mobile device is rooted;

notifying an administrator that the uncategorized mobile device comprises a rooted mobile device.

11. A system for categorizing mobile devices as rooted, the system comprising:

a gathering module, stored in memory, that gathers a set of metadata describing a plurality of rooted mobile devices that have been modified to allow a user to alter protected systems and an additional set of metadata describing a plurality of unrooted mobile devices that have not been modified to allow the user to alter the protected systems;

a comparison module, stored in memory, that compares the set of metadata with the additional set of metadata by processing the metadata and the additional set of metadata through a series of classifiers comprising a machine-learning algorithm and a linear regression algorithm to determine at least one feature that differentiates the rooted mobile devices from the unrooted mobile devices, wherein the at least one feature comprises an application category including one or more malware application types and one or more applications used for potentially malicious purposes;

a determination module, stored in memory, that determines whether the at least one feature is present in metadata that describes an uncategorized mobile device;

a categorization module, stored in memory, that:

categorizes the uncategorized mobile device as a rooted mobile device based on a total weighted value of the at least one feature present in the metadata exceeding a predetermined threshold; and

protects a secure network against the one or more malware application types and the one or more applications used for potentially malicious purposes potentially present on the rooted device by disallowing the rooted mobile device from connecting to the secure network; and

at least one physical processor configured to execute the gathering module, the comparison module, the determination module, and the categorization module.

12. The system of claim 11 , wherein the secure network is governed by a security policy that prohibits rooted devices from connecting to the secure network.

13. The system of claim 12 , wherein the categorization module:

categorizes an additional mobile device as an unrooted mobile device based on the absence of the at least one feature in metadata that describes the additional mobile device; and

allows the unrooted mobile device to connect to the secure network.

14. The system of claim 11 , wherein the determination module determines whether the at least one feature is present in the metadata that describes the uncategorized mobile device by determining, on the uncategorized mobile device, whether the at least one feature is present.

15. The system of claim 11 , wherein the determination module determines whether the at least one feature is present in the metadata that describes the uncategorized mobile device by determining, at a server, whether the at least one feature is present.

16. The system of claim 11 , wherein the one or more applications used for potentially malicious purposes installed on the uncategorized mobile device.

17. The system of claim 11 , wherein the at least one feature comprises a pattern of application installation behavior on the uncategorized mobile device that comprises at least one of:

a number of times applications have been installed on the uncategorized mobile device;

a number of times applications have been uninstalled on the uncategorized mobile device;

a number of times applications have been updated on the uncategorized mobile device;

an average time between two consecutive installations or uninstallations of applications on the uncategorized mobile device.

18. A non-transitory computer-readable medium comprising one or more computer-readable instructions that, when executed by at least one processor of a computing device, cause the computing device to:

gather a set of metadata describing a plurality of rooted mobile devices that have been modified to allow a user to alter protected systems and an additional set of metadata describing a plurality of unrooted mobile devices that have not been modified to allow the user to alter the protected systems;

compare the set of metadata with the additional set of metadata by processing the metadata and the additional set of metadata through a series of classifiers comprising a machine-learning algorithm and a linear regression algorithm to determine at least one feature that differentiates the rooted mobile devices from the unrooted mobile devices, wherein the at least one feature comprises an application category including one or more malware application types and one or more applications used for potentially malicious purposes;

determine whether the at least one feature is present in metadata that describes an uncategorized mobile device;

categorize the uncategorized mobile device as a rooted mobile device based on a total weighted value of the at least one feature present in the metadata exceeding a predetermined threshold; and

protect a secure network against the one or more malware application types and the one or more applications used for potentially malicious purposes potentially present on the rooted device by disallowing the rooted mobile device from connecting to the secure network.

19. The non-transitory computer-readable medium of claim 18 , wherein the secure network is governed by a security policy that prohibits rooted devices from connecting to the secure network.

20. The non-transitory computer-readable medium of claim 18 , wherein the one or more computer-readable instructions further cause the computing device to:

categorize an additional mobile device as an unrooted mobile device based on the absence of the at least one feature in metadata that describes the additional mobile device; and

allow the unrooted mobile device to connect to the secure network.

Assignments (6)