Leveraging Adversarial Samples for Enhanced Classification of Malicious and Evasive PDF Files
Source abstract
Abstract
The Portable Document Format (PDF) is considered one of the most popular formats due to its flexibility and portability across platforms. Although people have used machine learning techniques to detect malware in PDF files, the problem with these models is their weak resistance against evasion attacks, which constitutes a major security threat. The goal of this study is to introduce three machine learning-based systems that enhance malware detection in the presence of evasion attacks by substantially relying on evasive data to train malware and evasion detection models. To evaluate the robustness of the proposed systems, we used two testing datasets, a real dataset containing around 100,000 PDF samples and an evasive dataset containing 500,000 samples that we generated. We compared the results of the proposed systems to a baseline model that was not adversarially trained. When tested against the evasive dataset, the proposed systems provided an increase of around 80% in the f1-score compared to the baseline. This proves the value of the proposed approaches towards the ability to deal with evasive attacks.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Advanced Malware Detection Techniques · Adversarial Robustness in Machine Learning · Network Security and Intrusion Detection
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Electrical and Computer Engineering, American University of Beirut, Beirut 1107-2020, Lebanon
Funding
Maroun Semaan Faculty of Engineering and Architecture (MSFEA) at the American University of Beirut (AUB)
References (49)
Show source-provided references
1. The recent trends in cyber security: A review
2. A comprehensive review study of cyber-attacks and cyber security; Emerging trends and recent developments10.1016/j.egyr.2021.08.126 ↗
3. A Comprehensive Review on Malware Detection Approaches10.1109/ACCESS.2019.2963724 ↗
4. Blonce, A., Filiol, E., and Frayssignes, L. (2008, January 24–28). Portable Document Format (PDF) Security Analysis and Malware Threats. Proceedings of the Europe BlackHat 2008 Conference, Amsterdam, The Netherlands.
5. Fleury, N., Dubrunquez, T., and Alouani, I. (2021). PDF-Malware: An Overview on Threats, Detection and Evasion Attacks. arXiv.
6. Iwamoto, M., Oshima, S., and Nakashima, T. (2016, January 6–8). A Study of Malicious PDF Detection Technique. Proceedings of the 2016 10th International Conference on Complex, Intelligent, and Software Intensive Systems (CISIS), Fukuoka, Japan.10.1109/CISIS.2016.45 ↗
7. Digital Investigation of PDF Files: Unveiling Traces of Embedded Malware10.1109/MSEC.2018.2875879 ↗
8. Torres, J., and De los Santos, S. (2018, January 22–24). Malicious PDF Documents Detection using Machine Learning Techniques—A Practical Approach with Cloud Computing Applications. Proceedings of the 4th International Conference on Information Systems Security and Privacy, Funchal, Portugal.10.5220/0006609503370344 ↗
9. Issakhani, M., Victor, P., Tekeoglu, A., and Lashkari, A. (2022, January 9–11). PDF Malware Detection based on Stacking Learning. Proceedings of the 8th International Conference on Information Systems Security and Privacy, Online.10.5220/0010908400003120 ↗
10. Towards Adversarial Malware Detection: Lessons Learned from PDF-based Attacks10.1145/3332184 ↗
11. Biggio, B., Corona, I., Maiorca, D., Nelson, B., Šrndić, N., Laskov, P., Giacinto, G., and Roli, F. (2013). Proceedings of the Machine Learning and Knowledge Discovery in Databases, Prague, Czech Republic, 23–27 September 2013, Springer. Lecture Notes in Computer Science.
12. Zhang, J. (2018). MLPdf: An Effective Machine Learning Based Approach for PDF Malware Detection. arXiv.
13. Zhang, J. (2019). Machine Learning With Feature Selection Using Principal Component Analysis for Malware Detection: A Case Study. arXiv.
14. Khorshidpour, Z., Hashemi, S., and Hamzeh, A. (2016, January 12–15). Learning a Secure Classifier against Evasion Attack. Proceedings of the 2016 IEEE 16th International Conference on Data Mining Workshops (ICDMW), Barcelona, Spain.10.1109/ICDMW.2016.0049 ↗
15. Boosting training for PDF malware classifier via active learning10.1002/int.22451 ↗
16. Xu, W., Qi, Y., and Evans, D. (2016, January 21–24). Automatically Evading Classifiers. Proceedings of the Network and Distributed System Security Symposium, San Diego, CA, USA.
17. Maiorca, D., Corona, I., and Giacinto, G. (2013, January 8–10). Looking at the bag is not enough to find the bomb: An evasion of structural methods for malicious PDF files detection. Proceedings of the 8th ACM SIGSAC Symposium on Information, Computer and Communications Security, Hangzhou, China.10.1145/2484313.2484327 ↗
18. Šrndić, N., and Laskov, P. (2014, January 18–21). Practical Evasion of a Learning-Based Classifier: A Case Study. Proceedings of the 2014 IEEE Symposium on Security and Privacy, San Jose, CA, USA.10.1109/SP.2014.20 ↗
19. Carmony, C., Zhang, M., Hu, X., Vasisht Bhaskar, A., and Yin, H. (2016, January 21–24). Extract Me If You Can: Abusing PDF Parsers in Malware Detectors. Proceedings of the 2016 Network and Distributed System Security Symposium, San Diego, CA, USA.10.14722/ndss.2016.23483 ↗
20. Mila (2022, April 18). 16,800 Clean and 11,960 Malicious Files for Signature Testing and Research. Contagio Dataset. Available online: http://contagiodump.blogspot.com/2013/03/16800-clean-and-11960-malicious-files.html.
21. VirusTotal (2022, April 23). Virus Total Home Page. Available online: https://www.virustotal.com/gui/home/upload.
22. Smutz, C., and Stavrou, A. (2012, January 3–7). Malicious PDF detection using metadata and structural features. Proceedings of the 28th Annual Computer Security Applications Conference—ACSAC ’12, Orlando, FL, USA.10.1145/2420950.2420987 ↗
23. Xu, L., Skoularidou, M., Cuesta-Infante, A., and Veeramachaneni, K. (2019). Modeling Tabular data using Conditional GAN. arXiv.
24. Patki, N., Wedge, R., and Veeramachaneni, K. (2016, January 17–19). The Synthetic Data Vault. Proceedings of the 2016 IEEE International Conference on Data Science and Advanced Analytics (DSAA), Montreal, QC, Canada.10.1109/DSAA.2016.49 ↗
25. (2008). Document Management—Portable Document Format—Part 1: PDF 1.7 (Standard No. ISO:32000-1:2008). Available online: https://www.iso.org/cms/render/live/en/sites/isoorg/contents/data/standard/05/15/51502.html.
26. Tzermias, Z., Sykiotakis, G., Polychronakis, M., and Markatos, E.P. (2011, January 11). Combining static and dynamic analysis for the detection of malicious documents. Proceedings of the Fourth European Workshop on System Security, Salzburg, Austria.10.1145/1972551.1972555 ↗
27. (2006). Adobe® PDF (Portable Document Format) 1.7 Reference.
28. Corona, I., Maiorca, D., Ariu, D., and Giacinto, G. (2014, January 7). Lux0R: Detection of Malicious PDF-embedded JavaScript code through Discriminant Analysis of API References. Proceedings of the 2014 Workshop on Artificial Intelligent and Security Workshop, Scottsdale, AZ, USA.10.1145/2666652.2666657 ↗
29. Munson, M., and Cross, J. (2011). Deep PDF Parsing to Extract Features for Detecting Embedded Malware, OSTI.10.2172/1030303 ↗
30. Malicious PDF Documents Explained10.1109/MSP.2011.14 ↗
31. Croce, F., and Hein, M. (2020). Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attacks. arXiv.
32. Brendel, W., Rauber, J., and Bethge, M. (2018). Decision-Based Adversarial Attacks: Reliable Attacks Against Black-Box Machine Learning Models. arXiv.
33. Brendel, W., Rauber, J., Kümmerer, M., Ustyuzhaninov, I., and Bethge, M. (2019). Accurate, reliable and fast robustness evaluation. arXiv.
34. Moosavi-Dezfooli, S.M., Fawzi, A., and Frossard, P. (2016). DeepFool: A simple and accurate method to fool deep neural networks. arXiv.10.1109/CVPR.2016.282 ↗
35. Chen, P.Y., Sharma, Y., Zhang, H., Yi, J., and Hsieh, C.J. (2018). EAD: Elastic-Net Attacks to Deep Neural Networks via Adversarial Examples. arXiv.10.1609/aaai.v32i1.11302 ↗
36. Chen, J., Jordan, M.I., and Wainwright, M.J. (2020). HopSkipJumpAttack: A Query-Efficient Decision-Based Attack. arXiv.10.1109/SP40000.2020.00045 ↗
37. Wallach, H., Larochelle, H., Beygelzimer, A., d’Alché-Buc, F., Fox, E., and Garnett, R. (2019). Proceedings of the Advances in Neural Information Processing Systems, Curran Associates, Inc.
38. Papernot, N., McDaniel, P., Wu, X., Jha, S., and Swami, A. (2016, January 22–26). Distillation as a Defense to Adversarial Perturbations Against Deep Neural Networks. Proceedings of the 2016 IEEE Symposium on Security and Privacy (SP), San Jose, CA, USA.10.1109/SP.2016.41 ↗
39. Xu, W., Evans, D., and Qi, Y. (2018, January 18–21). Feature Squeezing: Detecting Adversarial Examples in Deep Neural Networks. Proceedings of the 2018 Network and Distributed System Security Symposium, Internet Society, San Diego, CA, USA.10.14722/ndss.2018.23198 ↗
40. Papernot, N., McDaniel, P., Sinha, A., and Wellman, M.P. (2018, January 24–26). SoK: Security and Privacy in Machine Learning. Proceedings of the 2018 IEEE European Symposium on Security and Privacy (EuroS&P), London, UK.10.1109/EuroSP.2018.00035 ↗
41. Papernot, N., McDaniel, P., and Goodfellow, I. (2016). Transferability in Machine Learning: From Phenomena to Black-Box Attacks using Adversarial Samples. arXiv.
42. Demontis, A., Melis, M., Pintor, M., Jagielski, M., Biggio, B., Oprea, A., Nita-Rotaru, C., and Roli, F. (2019). Why Do Adversarial Attacks Transfer? Explaining Transferability of Evasion and Poisoning Attacks. arXiv.
43. Implementation of Malware Analysis using Static and Dynamic Analysis Method
44. Embedded Malware Detection Using Markov n-Grams10.1007/978-3-540-70542-0_5 ↗
45. Smutz, C., and Stavrou, A. (2016, January 21–24). When a Tree Falls: Using Diversity in Ensemble Classifiers to Identify Evasion in Malware Detectors. Proceedings of the 2016 Network and Distributed System Security Symposium, San Diego, CA, USA.10.14722/ndss.2016.23078 ↗
46. Goodfellow, I.J., Shlens, J., and Szegedy, C. (2015). Explaining and Harnessing Adversarial Examples. arXiv.
47. Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., and Rabinovich, A. (2014). Going Deeper with Convolutions. arXiv.10.1109/CVPR.2015.7298594 ↗
48. Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., Ye, Q., and Liu, T.Y. (2017). Proceedings of the Advances in Neural Information Processing Systems, Long Beach, CA, USA, 4–9 December 2017, Curran Associates, Inc.
49. Scikit-learn: Machine Learning in Python