Enhancing State-of-the-art Classifiers with API Semantics to Detect Evolved Android Malware
The DOI matches its Crossref record; 33 of the 33 references it cites were found in a registry or an index.
Abstract
Machine learning (ML) classifiers have been widely deployed to detect Android malware, but at the same time the application of ML classifiers also faces an emerging problem. The performance of such classifiers degrades---or called ages---significantly over time given the malware evolution. Prior works have proposed to use retraining or active learning to reverse and improve aged models. However, the underlying classifier itself is still blind, unaware of malware evolution. Unsurprisingly, such evolution-insensitive retraining or active learning comes at a price, i.e., the labeling of tens of thousands of malware samples and the cost of significant human efforts. In this paper, we propose the first framework, called APIGraph, to enhance state-of-the-art malware classifiers with the similarity information among evolved Android malware in terms of semantically-equivalent or similar API usages, thus naturally slowing down classifier aging. Our evaluation shows that because of the slow-down of classifier aging, APIGraph saves significant amounts of human efforts required by active learning in labeling new malware samples.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
(2008) · Journal of Machine Learning Research · Cited by 33,051
(2013) · Cited by 5,726
Freebase: a collaboratively created graph database for structuring human knowledge
Bollacker, K., Evans, C., Paritosh, P., Sturge, T., & Taylor, J. (2008) · Proceedings of the 2008 ACM SIGMOD international conference on Management of data · doi:10.1145/1376616.1376746 · Cited by 4,852 · cited 1×
“Knowledge graphs [7, 17] have been successfully constructed and applied to in many real-world tasks, such as extracting information and answering questions.”
BackgroundKnowledge Graph Embedding by Translating on Hyperplanes
Wang, Z., Zhang, J., Feng, J., & Chen, Z. (2014) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v28i1.8870 · Cited by 3,915
Drebin: Effective and Explainable Detection of Android Malware in Your Pocket
Arp, D., Spreitzenbarth, M., Hübner, M., Gascon, H., & Rieck, K. (2014) · Proceedings 2014 Network and Distributed System Security Symposium · doi:10.14722/ndss.2014.23247 · Cited by 2,331 · cited 19×
“In terms of the API feature, Drebin considers a selected set of restricted and suspicious APIs that can access to critical and sensitive data or resources.”
BackgroundMethodInfluentialSyakur, M. A., Khotimah, B. K., Rochman, E. M. S., & Satoto, B. D. (2018) · IOP Conference Series: Materials Science and Engineering · doi:10.1088/1757-899x/336/1/012017 · Cited by 1,181 · cited 2×
“Following the Elbow method [43], we choose 2,000 as the total cluster number.”
MethodProceedings of the 26th AAAI Conference on Artificial Intelligence (AAAI'12)
Hoffmann, J., & Selman, B. (2012) · HAL · Cited by 754
DroidAPIMiner: Mining API-Level Features for Robust Malware Detection in Android
Aafer, Y., Du, W., & Yin, H. (2013) · Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering · doi:10.1007/978-3-319-04283-1_6 · Cited by 686 · cited 3×
“APIs are a popular type of features widely adopted by many other existing malware classifiers [1, 11, 12, 22], mainly because APIs are essential in implementing malware functionalities.”
BackgroundMethodPScout: analyzing the Android permission specification
Au, K. W. Y., Zhou, Y. F., Huang, Z., & Lie, D. (2012) · Proceedings of the 2012 ACM conference on Computer and communications security · doi:10.1145/2382196.2382222 · Cited by 644 · cited 1×
“Note that APIGraph also extracts uses_permission relations from two API-permission mappings generated by existing works [4, 5] to complement the relations extracted from API documents, because such information in the Android API documents may be incomplete.”
Adversarial Examples for Malware Detection
Grosse, K., Papernot, N., Manoharan, P., Backes, M., & McDaniel, P. (2017) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-66399-9_4 · Cited by 542 · cited 4×
“We apply APIGraph upon four prior Android malware classifiers, namely MamaDroid [32], DroidEvolver [49], Drebin [3], and Drebin-DL [18], and evaluate them using a dataset created by ourselves following existing guidelines [39], which contains more than 322K Android apps ranging from 2012 to 2018.”
BackgroundMethodInfluentialDeep Android Malware Detection
McLaughlin, N., Martinez del Rincon, J., Kang, B., Yerima, S., Miller, P., Sezer, S., Safaei, Y., Trickel, E., Zhao, Z., Doupé, A., et al. (2017) · Proceedings of the Seventh ACM on Conference on Data and Application Security and Privacy · doi:10.1145/3029806.3029823 · Cited by 493 · cited 1×
“Machine learning (ML) classifiers are widely used in practice to detect Android malware [1, 3, 12, 18, 19, 29, 32, 33, 37, 49, 50, 52] and have achieved astonishing performance.”
BackgroundSemantics-Aware Android Malware Classification Using Weighted Contextual API Dependency Graphs
Zhang, M., Duan, Y., Yin, H., & Zhao, Z. (2014) · Proceedings of the 2014 ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/2660267.2660359 · Cited by 453 · cited 1×
“Machine learning (ML) classifiers are widely used in practice to detect Android malware [1, 3, 12, 18, 19, 29, 32, 33, 37, 49, 50, 52] and have achieved astonishing performance.”
BackgroundSelf-citationTranscend: Detecting Concept Drift in Malware Classification Models
Jordaney, R., Sharad, K., Dash, S. K., Wang, Z., Papini, D., Nouretdinov, I., & Cavallaro, L. (2017) · Research Portal (King's College London) · Cited by 447
Ding, S. H. H., Fung, B. C. M., & Charland, P. (2019) · 2019 IEEE Symposium on Security and Privacy (SP) · doi:10.1109/sp.2019.00003 · Cited by 438 · cited 1×
“Other solutions such as API usage mining with large-scale market apps [14, 38] may also generate similar relations.”
BackgroundMalDozer: Automatic framework for android malware detection using deep learning
Karbab, E. B., Debbabi, M., Derhab, A., & Mouheb, D. (2018) · Digital Investigation · doi:10.1016/j.diin.2018.01.007 · Cited by 403 · cited 2×
“APIs are a popular type of features widely adopted by many other existing malware classifiers [1, 11, 12, 22], mainly because APIs are essential in implementing malware functionalities.”
BackgroundMethodDeep Ground Truth Analysis of Current Android Malware
Wei, F., Li, Y., Roy, S., Ou, X., & Zhou, W. (2017) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-60876-1_12 · Cited by 398 · cited 1×
“It is worth noting that our enhancement of classifiers depends on the appearance year of the target apps—that is, if the target apps of the classifiers are from the year 2012, our enhancement will be using the API relation graph of API Level 18, because this is the latest API level in 2012.”
MethodExecute This! Analyzing Unsafe and Malicious Dynamic Code Loading in Android Applications
Poeplau, S., Fratantonio, Y., Bianchi, A., Kruegel, C., & Vigna, G. (2014) · Proceedings 2014 Network and Distributed System Security Symposium · doi:10.14722/ndss.2014.23328 · Cited by 265 · cited 1×
“This is an orthogonal problem to what has been studied in APIGraph and one should refer to existing works [6, 26, 40, 53] for solutions.”
BackgroundHou, S., Saas, A., Chen, L., & Ye, Y. (2016) · 2016 IEEE/WIC/ACM International Conference on Web Intelligence Workshops (WIW) · doi:10.1109/wiw.2016.040 · Cited by 225 · cited 1×
“Machine learning (ML) classifiers are widely used in practice to detect Android malware [1, 3, 12, 18, 19, 29, 32, 33, 37, 49, 50, 52] and have achieved astonishing performance.”
BackgroundMARVIN: Efficient and Comprehensive Mobile App Classification through Static and Dynamic Analysis
Lindorfer, M., Neugschwandtner, M., & Platzer, C. (2015) · 2015 IEEE 39th Annual Computer Software and Applications Conference · doi:10.1109/compsac.2015.103 · Cited by 205 · cited 1×
“Machine learning (ML) classifiers are widely used in practice to detect Android malware [1, 3, 12, 18, 19, 29, 32, 33, 37, 49, 50, 52] and have achieved astonishing performance.”
BackgroundStormDroid: A Streaminglized Machine Learning-Based System for Detecting Android Malware
Chen, S., Xue, M., Tang, Z., Xu, L., & Zhu, H. (2016) · ACM Asia Conference on Computer and Communications Security (ASIA CCS) · doi:10.1145/2897845.2897860 · Cited by 197 · cited 1×
“Machine learning (ML) classifiers are widely used in practice to detect Android malware [1, 3, 12, 18, 19, 29, 32, 33, 37, 49, 50, 52] and have achieved astonishing performance.”
BackgroundPatterns of Knowledge in API Reference Documentation
Maalej, W., & Robillard, M. P. (2013) · IEEE Transactions on Software Engineering · doi:10.1109/tse.2013.12 · Cited by 183
AppContext: Differentiating Malicious and Benign Mobile App Behaviors Using Context
Yang, W. M., Xiao, X., Andow, B., Li, S., Xie, T., & Enck, W. (2015) · ACM/IEEE International Conference on Software Engineering: Companion Proceedings (ICSE-Companion) · doi:10.1109/icse.2015.50 · Cited by 170
DroidEvolver: Self-Evolving Android Malware Detection System
Xu, K., Li, Y., Deng, R., Chen, K., & Xu, J. (2019) · 2019 IEEE European Symposium on Security and Privacy (EuroS&P) · doi:10.1109/eurosp.2019.00014 · Cited by 160 · cited 4×
“We apply APIGraph upon four prior Android malware classifiers, namely MamaDroid [32], DroidEvolver [49], Drebin [3], and Drebin-DL [18], and evaluate them using a dataset created by ourselves following existing guidelines [39], which contains more than 322K Android apps ranging from 2012 to 2018.”
BackgroundMethodInfluentialFinding unknown malice in 10 seconds: mass vetting for new threats at the Google-play scale
Chen, K., Wang, P., Lee, Y., Wang, X., Zhang, N., Huang, H., Zou, W., & Liu, P. (2015) · USENIX Security Symposium · Cited by 160
Exploring API Embedding for API Usages and Applications
Nguyen, T. D., Nguyen, A. T., Phan, H. D., & Nguyen, T. N. (2017) · 2017 IEEE/ACM 39th International Conference on Software Engineering (ICSE) · doi:10.1109/icse.2017.47 · Cited by 159
Assessing and Improving Malware Detection Sustainability through App Evolution Studies
Cai, H. (2020) · ACM Transactions on Software Engineering and Methodology · doi:10.1145/3371924 · Cited by 136 · cited 1×
“This problem is defined as model aging or similar concepts like time decay [39], model degradation [24], and deterioration [9] in the literature.”
BackgroundDexHunter: Toward Extracting Hidden Code from Packed Android Applications
Zhang, Y., Luo, X., & Yin, H. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-24177-7_15 · Cited by 136
Self-citationEveDroid: Event-Aware Android Malware Detection Against Model Degrading for IoT Devices
Lei, T., Qin, Z., Wang, Z., Li, Q. L., & Ye, D. (2019) · IEEE Internet of Things Journal · doi:10.1109/jiot.2019.2909745 · Cited by 121 · cited 1×
“This problem is defined as model aging or similar concepts like time decay [39], model degradation [24], and deterioration [9] in the literature.”
BackgroundEuphony: Harmonious Unification of Cacophonous Anti-Virus Vendor Labels for Android Malware
Hurier, M., Suárez‐Tangil, G., Dash, S. K., Bissyandé, T. F., Traon, Y. L., Klein, J., & Cavallaro, L. (2017) · International Conference on Mining Software Repositories (MSR) · doi:10.1109/msr.2017.57 · Cited by 104 · cited 1×
“Figure 9 shows a subgraph of the visualization, in which those APIs in our motivating example (Figure 1) are clearly separated into different clusters.”
BackgroundImproving API Caveats Accessibility by Mining API Caveats Knowledge Graph
Li, H., Li, S., Sun, J., Xing, Z., Peng, X., Liu, M., & Zhao, X. (2018) · 2018 IEEE International Conference on Software Maintenance and Evolution (ICSME) · doi:10.1109/icsme.2018.00028 · Cited by 97
Static Analysis of Implicit Control Flow: Resolving Java Reflection and Android Intents (T)
Barros, P., Just, R., Millstein, S., Vines, P., Dietl, W., d’Amorim, M., & Ernst, M. D. (2015) · IEEE/ACM International Conference on Automated Software Engineering (ASE) · doi:10.1109/ase.2015.69 · Cited by 82 · cited 1×
“This is an orthogonal problem to what has been studied in APIGraph and one should refer to existing works [6, 26, 40, 53] for solutions.”
BackgroundAdaptive and scalable Android malware detection through online learning
Narayanan, A. S., Liu, Y., Chen, L., & Liu, J. (2016) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn.2016.7727508 · Cited by 81 · cited 2×
“Machine learning (ML) classifiers are widely used in practice to detect Android malware [1, 3, 12, 18, 19, 29, 32, 33, 37, 49, 50, 52] and have achieved astonishing performance.”
BackgroundChen, L., Hou, S., Ye, Y., & Xu, S. (2018) · IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM) · doi:10.1109/asonam.2018.8508284 · Cited by 33
Metadata
- Authors
- Zhang, X., Zhang, Y., Zhong, M., Ding, D., Cao, Y., Zhang, Y., Zhang, M., & Yang, M.
- Venue
- Proceedings of the 2020 ACM SIGSAC Conference on Computer and Communications Security
- Publisher
- ACM
- Published
- 2020-10-30
- Pages
- 757-770
- OpenAlex
- W3097730806
- Cited by
- 185
- References
- 33
- FWCI
- 16.40
- Type
- conference paper
- Language
- en
- Open access
- closed
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.