Research on unsupervised feature learning for Android malware detection based on Restricted Boltzmann Machines

Liu, Zhen; Wang, Ruoyu; Japkowicz, Nathalie; Tang, Deyu; Zhang, Wenbin; Zhao, Jie

doi:10.1016/j.future.2021.02.015

Cited by 49 publications

(14 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nowadays, the application of cloud database in the field of modern education has already become a trend [1,2]. e cloud database platform has the advantages of large capacity, high reliability, strong scalability, strong compatibility, and high transmission efficiency.…”

Section: Introductionmentioning

confidence: 99%

Mobile Learning Platform in Cloud Computing with Information Security and Android System

Guan¹

2022

Security and Communication Networks

View full text Add to dashboard Cite

In order to meet the real-time communication between teachers and students and improve students' interest and efficiency in learning, this paper designs a mobile learning communication system based on information security and Android system with cloud computing. By using the method of combining database records and web logs to mine the user's browsing records and behaviors, these implicit user behaviors are transformed into explicit user evaluations of the project. Then, the cosine similarity calculation method is used to calculate users and the similarity between the users. The users are clustered by the K-means clustering method, so that the users are automatically divided into several user clusters according to their behaviors. Finally, the user's nearest neighbor score is used to predict the pair. Based on the above method, a mobile learning communication system based on Android is realized, and the system basically meets the functional needs of users. The development of this system not only promotes the mutual communication between students but also facilitates the students' learning, which has a certain promoting effect on the improvement of their academic performance.

show abstract

Section: Introductionmentioning

confidence: 99%

Mobile Learning Platform in Cloud Computing with Information Security and Android System

Guan¹

2022

Security and Communication Networks

View full text Add to dashboard Cite

show abstract

“…However, it is a shocking fact that the examined 26 Android malware detection frameworks using hybrid analysis, that only nine frameworks just consider few evasion techniques such as RiskRanker ( Grace et al, 2012 ) that has initiated the issue in 2012, Mobile-Sandbox ( Hoffmann et al, 2016 ), Marvin ( Lindorfer, Neugschwandtner & Platzer, 2015 ). Recently some hybrid analysis based detection Puerta ( de la Puerta et al, 2019 ), Surendrean ( Surendran, Thomas & Emmanuel, 2020 ), Lu ( Lu et al, 2020 ), Dhalaria ( Dhalaria & Gandotra, 2021 ), Zhu ( Zhu et al, 2021 ), Nawaz ( Nawaz, 2021 ), Liu ( Liu et al, 2021 ), PNSDroid ( Kandukuru & Sharma, 2018 ), Bacci ( Bacci et al, 2018 ), DAMBA ( Zhang et al, 2020 ) has highlighted the complex evasions detection resiliency issue in their research literature; however, the proposed malware detection methods and experiments of excluded the obfuscated malware from their evaluation sheets.…”

Section: Discussionmentioning

confidence: 99%

The rise of obfuscated Android malware and impacts on detection methods

Elsersy¹,

Feizollah²,

Anuar³

2022

PeerJ Computer Science

View full text Add to dashboard Cite

The various application markets are facing an exponential growth of Android malware. Every day, thousands of new Android malware applications emerge. Android malware hackers adopt reverse engineering and repackage benign applications with their malicious code. Therefore, Android applications developers tend to use state-of-the-art obfuscation techniques to mitigate the risk of application plagiarism. The malware authors adopt the obfuscation and transformation techniques to defeat the anti-malware detections, which this paper refers to as evasions. Malware authors use obfuscation techniques to generate new malware variants from the same malicious code. The concern of encountering difficulties in malware reverse engineering motivates researchers to secure the source code of benign Android applications using evasion techniques. This study reviews the state-of-the-art evasion tools and techniques. The study criticizes the existing research gap of detection in the latest Android malware detection frameworks and challenges the classification performance against various evasion techniques. The study concludes the research gaps in evaluating the current Android malware detection framework robustness against state-of-the-art evasion techniques. The study concludes the recent Android malware detection-related issues and lessons learned which require researchers’ attention in the future.

show abstract

“…Hei et al [ 18 ] used the system API call sequence as the basis for determining malware, which can be able to restore the deformed call software dynamically, partially solving the software deformation problem. Liu et al [ 19 ] used the program data stored in the PE format file to analyze and obtain the DLL file from which the program calls the system. Huang et al [ 20 ] proposed an inference tree-based analysis method.…”

Section: Related Workmentioning

confidence: 99%

A Study on the Application of Distributed System Technology-Guided Machine Learning in Malware Detection

Shi

Guo²,

Liu³

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

In recent years, with the development of information technology, the Internet has become an essential tool for human daily life. However, as the popularity and scale of the Internet continue to expand, malware has also emerged as an increasingly widespread trend, and its development has brought many negative impacts to the society. As the number of types of malware is getting enormous, the attacks are constantly updated, and at the same time, the spread is very fast, causing more and more damage to the network, the requirements and standards for malware detection are constantly rising. How to effectively detect malware is a research trend; in order to tackle the new needs and problems arising from the development of malware, this paper proposes to guide machine learning algorithms to implement malware detection in a distributed environment: firstly, each detection node in the distributed network performs anomaly detection on the captured software information and data, then performs feature analysis to discover unknown malware and obtain its samples, updates the new malware features to all feature detection nodes in the whole distributed network, and trains the random forest-based machine learning algorithm for malware classification and detection, thus completing the global response processing capability for malware. By building a distributed system framework, the global capture capability of malware detection is enhanced to robustly respond to the increasing and rapid spread of malware, and machine learning algorithms are integrated into it to achieve effective detection of malware. Extended experiments on the Ember 2017 and Ember 2018 databases show that our proposed approach achieves advanced performance and effectively addresses the problem of malware detection.

show abstract

Research on unsupervised feature learning for Android malware detection based on Restricted Boltzmann Machines

Cited by 49 publications

References 41 publications

Mobile Learning Platform in Cloud Computing with Information Security and Android System

Mobile Learning Platform in Cloud Computing with Information Security and Android System

The rise of obfuscated Android malware and impacts on detection methods

A Study on the Application of Distributed System Technology-Guided Machine Learning in Malware Detection

Contact Info

Product

Resources

About