Low-Resource Footprint, Data-Driven Malware Detection on Android

Aonzo, Simone; Merlo, Alessio; Migliardi, Mauro; Oneto, Luca; Palmieri, Francesco

doi:10.1109/tsusc.2017.2774184

Cited by 20 publications

(20 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many studies have shown that feature selection approaches that select good feature subset will have significant impact on reducing the complexity in processing by eliminating unimportant features and enhance the performance of the learning models [24,30]. For example, Aonzo et al [30] demonstrated that small number of features are enough for a very good classification. They considered the most significant features extensively used in prior research.…”

Section: Related Workmentioning

confidence: 99%

“…Although choosing a subset of features from the original features is a combinatorial problem, many suboptimal heuristics have been put forward and used in various domains, which include the chi-squared based feature subset selection [ 7 , 8 , 10 ], the analysis of variance (ANOVA) [ 7 , 8 , 10 ], mutual information [ 7 , 23 , 29 ] and information gain [ 18 , 25 , 26 , 27 ]. Many studies have shown that feature selection approaches that select good feature subset will have significant impact on reducing the complexity in processing by eliminating unimportant features and enhance the performance of the learning models [ 24 , 30 ]. For example, Aonzo et al [ 30 ] demonstrated that small number of features are enough for a very good classification.…”

Section: Related Workmentioning

confidence: 99%

“…The prior works are either focused on non-malware related fields [ 19 , 30 , 31 ] or compare approaches that transform the original features against approaches that does not [ 10 ] or use classification methods that perform embedded selection of the feature subsets such as random forests [ 41 ] and gradient boosting [ 39 ] or use them as a validation of a newly proposed feature subset selection methods [ 10 ]. As noted in [ 19 ], it is essential that the efficacy of the feature subset selection is examined and verified on different situations and platforms such as Android malware classification.…”

Section: Related Workmentioning

confidence: 99%

“…Android uses several permissions to guard users’ sensitive information from Android apps on the smartphone [ 30 , 43 ]. Each permission is associated with a specific operation such as accessing secure sections of the API, contact lists and camera.…”

Section: Android Malware Detection Frameworkmentioning

confidence: 99%

See 3 more Smart Citations

Feature Subset Selection for Malware Detection in Smart IoT Platforms

Abawajy

Darem

Alhashmi

2021

Sensors

View full text Add to dashboard Cite

Malicious software (“malware”) has become one of the serious cybersecurity issues in Android ecosystem. Given the fast evolution of Android malware releases, it is practically not feasible to manually detect malware apps in the Android ecosystem. As a result, machine learning has become a fledgling approach for malware detection. Since machine learning performance is largely influenced by the availability of high quality and relevant features, feature selection approaches play key role in machine learning based detection of malware. In this paper, we formulate the feature selection problem as a quadratic programming problem and analyse how commonly used filter-based feature selection methods work with emphases on Android malware detection. We compare and contrast several feature selection methods along several factors including the composition of relevant features selected. We empirically evaluate the predictive accuracy of the feature subset selection algorithms and compare their predictive accuracy and the execution time using several learning algorithms. The results of the experiments confirm that feature selection is necessary for improving accuracy of the learning models as well decreasing the run time. The results also show that the performance of the feature selection algorithms vary from one learning algorithm to another and no one feature selection approach performs better than the other approaches all the time.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Android Malware Detection Frameworkmentioning

confidence: 99%

See 2 more Smart Citations

Feature Subset Selection for Malware Detection in Smart IoT Platforms

Abawajy

Darem

Alhashmi

2021

Sensors

View full text Add to dashboard Cite

show abstract

“…Aonzo et al [28] developed an application available on the Google Play Store called BAdDroIds. They proposed a static analysis method using on-device machine learning that is supposedly efficient.…”

Section: Malware Detection Methodsmentioning

confidence: 99%

A Client/Server Malware Detection Model Based on Machine Learning for Android Devices

Fournier

Khoury

Pierre

2021

IoT

View full text Add to dashboard Cite

The rapid adoption of Android devices comes with the growing prevalence of mobile malware, which leads to serious threats to mobile phone security and attacks private information on mobile devices. In this paper, we designed and implemented a model for malware detection on Android devices to protect private and financial information, for the mobile applications of the ATISCOM project. This model is based on client/server architecture, to reduce the heavy computations on a mobile device by sending data from the mobile device to the server for remote processing (i.e., offloading) of the predictions. We then gradually optimized our proposed model for better classification of the newly installed applications on Android devices. We at first adopted Naive Bayes to build the model with 92.4486% accuracy, then the classification method that gave the best accuracy of 93.85% for stochastic gradient descent (SGD) with binary class (i.e., malware and benign), and finally the regression method with numerical values ranging from −100 to 100 to manage the uncertainty predictions. Therefore, our proposed model with random forest regression gives a good accuracy in terms of performance, with a good correlation coefficient, minimum computation time and the smallest number of errors for malware detection.

show abstract