Systems and methods for malware detection using localized machine learning
The disclosed computer-implemented method for malware detection using localized machine learning may include (i) generating a global score for a file using a global machine learning model, (ii) generating a localized score for the file using a localized machine learning model, (iii) determining that the file is malware using the global score, the localized score, and the local conviction threshold, and (iv) in response to determining that the file is malware, performing a security action to protect the computing device against malware. Various other methods, systems, and computer-readable media are also disclosed.
1. A computer-implemented method for malware detection using localized machine learning, at least a portion of the method being performed by a computing device comprising at least one processor, the method comprising:
generating, by a computing device of a computing environment, a global score for a file using a global machine learning model;
generating a localized score for the file using a localized machine learning model;
identifying a number of new static clusters of computing devices in the computing environment of the computing device in a time period;
determining that the number of the new static clusters is below an identified threshold;
determining to use the localized machine learning model;
determining that the file is malware using the global score, the localized score, and a local conviction threshold; and
in response to determining that the file is malware, performing a security action to protect the computing device against malware.
2. The computer-implemented method of claim 1 , wherein the global machine learning model is trained using a global dataset from a plurality of computing devices.
3. The computer-implemented method of claim 1 , wherein the localized machine learning model is generated by re-training the global machine learning model using a local dataset of the computing device.
4. The computer-implemented method of claim 1 , wherein the security action comprises:
displaying, to a user of the computing device, an alert indicating that the file is malware; and
removing the file from the computing device.
5. The computer-implemented method of claim 1 , further comprising:
calculating a false positive rate of the computing device; and
determining the local conviction threshold using the false positive rate of the computing device.
6. The computer-implemented method of claim 5 , further comprising:
recalculating the false positive rate of the computing device; and
adjusting the local conviction threshold using the recalculated false positive rate of the computing device.
7. The computer-implemented method of claim 5 , further comprising:
determining the false positive rate of the computing device by obtaining data from different malware detection systems and identification of a number of files incorrectly identified as malware.
8. A system for malware detection using localized machine learning, the system comprising:
at least one physical processor;
physical memory comprising computer-executable instructions that, when executed by the physical processor, cause the physical processor to:
generate a global score for a file using a global machine learning model of a computing device;
generate a localized score for the file using a localized machine learning model;
identify a number of new static clusters of computing devices in a computing environment of the computing device in a time period;
determine that the number of the new static clusters is below an identified threshold;
determine to use the localized machine learning model;
determine that the file is malware using the global score, the localized score, and a local conviction threshold; and
in response to determining that the file is malware, perform a security action to protect the computing device against malware.
9. The system of claim 8 , wherein the global machine learning model is trained using a global dataset from a plurality of computing devices.
10. The system of claim 8 , wherein the localized machine learning model is generated by re-training the global machine learning model using a local dataset of the computing device.
11. The system of claim 8 , wherein the security action comprises:
displaying, to a user of the computing device, an alert indicating that the file is malware; and
removing the file from the computing device.
12. The system of claim 8 , wherein the computer-executable instructions further cause the physical processor to:
calculate a false positive rate of the computing device; and
determine the local conviction threshold using the false positive rate of the computing device.
13. The system of claim 12 , wherein the computer-executable instructions further cause the physical processor to:
recalculate the false positive rate of the computing device; and
adjust the local conviction threshold using the recalculated false positive rate of the computing device.
14. The system of claim 12 , wherein the computer-executable instructions further cause the physical processor to:
determine the false positive rate of the computing device by obtaining data from different malware detection systems and identification of a number of files incorrectly identified as malware.
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:
generate a global score for a file using a global machine learning model;
generate a localized score for the file using a localized machine learning model;
identify a number of new static clusters of computing devices in a computing environment of the computing device in a time period;
determine that the number of the new static clusters is below an identified threshold;
determine to use the localized machine learning model;
determine that the file is malware using the global score, the localized score, and a local conviction threshold; and
in response to determining that the file is malware, perform a security action to protect the computing device against malware.
16. The non-transitory computer-readable medium of claim 15 , wherein the global machine learning model is trained using a global dataset from a plurality of computing devices.
17. The non-transitory computer-readable medium of claim 15 , wherein the localized machine learning model is generated by re-training the global machine learning model using a local dataset of the computing device.
18. The non-transitory computer-readable medium of claim 15 , wherein the security action comprises:
displaying, to a user of the computing device, an alert indicating that the file is malware; and
removing the file from the computing device.
19. The non-transitory computer-readable medium of claim 15 , wherein the one or more computer-executable instructions further cause the computing device to:
calculate a false positive rate of the computing device; and
determine the local conviction threshold using the false positive rate of the computing device.
20. The non-transitory computer-readable medium of claim 19 , wherein the one or more computer-executable instructions further cause the computing device to:
recalculate the false positive rate of the computing device; and
adjust the local conviction threshold using the recalculated false positive rate of the computing device.