System and method for detecting malware in documents
View Patent ↗In one embodiment, a method includes identifying, using one or more processors, a plurality of characteristics of a Portable Document Format (PDF) file. The method also includes determining, using the one or more processors, for each of the plurality of characteristics, a score corresponding to the characteristic. In addition, the method includes comparing, using the one or more processors, the determined scores to a first threshold. Based at least on the comparison of the determined scores to the first threshold, the method includes determining, using the one or more processors, that the PDF file is potential malware.
1. A method, comprising:
identifying, using one or more processors, a plurality of characteristics of a Portable Document Format (PDF) file, wherein the plurality of characteristics comprise: a time zone, metadata, language, font type, incorrect syntax, duplicate object numbers, and an embedded executable;
determining, using the one or more computer processors, for each characteristic of the plurality of characteristics, a score corresponding to the characteristic, wherein the score for each of the characteristics is determined based on the frequency with which each characteristic occurs in a set of PDF files known to be malware and the frequency with which each characteristic occurs in a set of PDF files known not to be malware;
comparing, using the one or more computer processors, the determined scores to a first threshold; and
based at least on the comparison of the determined scores to the first threshold, determining, using the one or more processors, that the PDF file is potential malware.
2. The method of claim 1 , wherein:
the method comprises comparing the determined scores to a second threshold; and
the PDF file is determined to be potential malware by determining that the determined scores are less than the second threshold and greater than the first threshold.
3. The method of claim 1 , further comprising sending, in response to determining that the PDF file is potential malware, the PDF file, the plurality of characteristics, and the determined scores to a human review module for review by a human analyst.
4. The method of claim 1 , wherein comparing the determined scores to the first threshold comprises comparing a weighted sum of the determined scores to the first threshold.
5. The method of claim 1 , wherein the plurality of characteristics comprises existence of Javascript.
6. The method of claim 1 , wherein the plurality of characteristics comprises the presence of markers between PDF objects.
7. The method of claim 1 , wherein the plurality of characteristics comprises multiple PDF objects having the same object number.
8. A system, comprising:
one or more computer processors configured to:
identify a plurality of characteristics of a Portable Document Format (PDF) file, wherein the plurality of characteristics comprise: a time zone, metadata, language, font type, incorrect syntax, duplicate object numbers, and an embedded executable;
determine for each characteristic of the plurality of characteristics, a score corresponding to the characteristic, wherein the score for each of the characteristics is determined based on the frequency with which each characteristic occurs in a set of PDF files known to be malware and the frequency with which each characteristic occurs in a set of PDF files known not to be malware;
compare the determined scores to a first threshold; and
based at least on the comparison of the determined scores to the first threshold, determine that the PDF file is potential malware.
9. The system of claim 8 , wherein:
the one or more processors are configured to compare the determined scores to a second threshold; and
the PDF file is determined to be potential malware by determining that the determined scores are less than the second threshold and greater than the first threshold.
10. The system of claim 8 , wherein the one or more processors are configured to send, in response to determining that the PDF file is potential malware, the PDF file, the plurality of characteristics, and the determined scores to a human review module for review by a human analyst.
11. The system of claim 8 , wherein the one or more processors are configured to compare the determined scores to the first threshold by comparing a weighted sum of the determined scores to the first threshold.
12. The system of claim 8 , wherein the plurality of characteristics comprises existence of Javascript.
13. The system of claim 8 , wherein the plurality of characteristics comprises the presence of markers between PDF objects.
14. The system of claim 8 , wherein the plurality of characteristics comprises multiple PDF objects having the same object number.
15. At least one non-transitory computer-readable medium comprising instructions that, when executed by one or more processors, are configured to:
identify a plurality of characteristics of a Portable Document Format (PDF) file, wherein the plurality of characteristics comprise: a time zone, metadata, language, font type, incorrect syntax, duplicate object numbers, and an embedded executable;
determine for each characteristic of the plurality of characteristics, a score corresponding to the characteristic, wherein the score for each of the characteristics is determined based on the frequency with which each characteristic occurs in a set of PDF files known to be malware and the frequency with which each characteristic occurs in a set of PDF files known not to be malware;
compare the determined scores to a first threshold; and
based at least on the comparison of the determined scores to the first threshold, determine that the PDF file is potential malware.
16. The at least one computer-readable medium of claim 15 , wherein:
the instructions are configured to compare the determined scores to a second threshold; and
the PDF file is determined to be potential malware by determining that the determined scores are less than the second threshold and greater than the first threshold.
17. The at least one computer-readable medium of claim 15 , wherein the instructions are configured to compare the determined scores to the first threshold by comparing a weighted sum of the determined scores to the first threshold.
18. The method of claim 1 , wherein comparing the determined scores to a first threshold includes comparing the determined scores to a first threshold, the first threshold based on a context of how the PDF file was received.
19. The system of claim 8 , wherein the first threshold is based on a context of how the PDF file was received.
20. The at least one computer-readable medium of claim 15 , wherein the instructions configured to compare the determined scores to a first threshold are further configured to compare the determined scores to a first threshold, the first threshold based on a context of how the PDF file was received.