Malware Detection in Android-Based Mobile Environments Using Optimum-Path Forest

Costa, Kelton Augusto Pontara da; Silva, Luis Alexandre da; Martins, Guilherme Brandão; Rosa, Guilherme Jordão de Magalhães; Pereira, Clayton Reginaldo; Papa, João Paulo

doi:10.1109/icmla.2015.72

“…Some work [12], [13], [15], [16], [17] considered action as features in Malicious application recognition. Shao [12] removed the number of exercises and different features to identify malignant applications.…”

Section: App Componentmentioning

confidence: 99%

“…Shao [12] removed the number of exercises and different features to identify malignant applications. Studies [13], [15], [17] further applied assistance and broadcast beneficiaries as features in malicious application discovery. Feldman et al [18] picked the recurrence of highly need recipients and manhandled administrations to recognize malicious applications.…”

Section: App Componentmentioning

confidence: 99%

Malicious Applications Detection in Android using Machine Learning

Asif¹,

Asif²,

Mubarik³

et al. 2022

Preprint

0

View full text Add to dashboard Cite

A huge number of applications available for Android-based smartphone devices have emerged over the past years. Due to which a huge number of malicious applications has been growing explosively. Many approaches have been proposed to ensure the security and quality of application in the markets. Usually, Machine Learning approaches are utilized in the classification process of malicious application detection. Calculating accurate results of characterizing applications behaviors, or other features, has a direct effect on the results with Machine Learning calculations. Android applications emerge so quickly. The behavior of current applications has gotten progressively malicious. The extraction of malware-infected features from applications is thus become a difficult task. According to our knowledge, a ton of features have been extricated in existing work however no survey has overviewed the features built for identifying malicious applications efficiently. In this paper, we will in general give an extensive review of such sort of work that identifies feature applications by describing various practices of uses with various kinds of features. In this survey we have discussed the following dimensions: extraction and selection of feature methods if any, methods of detection and evaluation performed. In light of our review, we notice the issues of investigating malware-affected features from applications, give the scientific categorization and demonstrate the future headings.

show abstract

“…The reason for this technique was to check whether the applications exceeded their normal dangers. Some studies (Wermke et al, 2019;Faruki et al, 2013;Zhao et al, 2016;Yang et al, 2014;Suarez-Tangil et al, 2017;Chen et al, 2014;Nath and Mehtre, 2014;Avdiienko et al, 2015;da Costa et al, 2016;Saracino et al, 2018;Vidas et al, 2014;Feldman et al, 2015;Wu et al, 2016;Lindorfer et al, 2015;Chen et al, 2017;Yang et al, 2014;Cen et al, 2015;Zhu et al, 2016;Idrees et al, 2017;Dam and Touili, 2017;Chen et al, 2016;Leeds et al, 2017;Lin et al, 2017) brought permissions just as some different features and used ML ways to deal with recognize malicious applications. This methodology-accomplished precision as over 94% (Wei et al, 2019), Because applying permission security is critical for attackers to achieve their hacking goals, permission is the most commonly used static feature in Android.…”

Section: Permissionmentioning

confidence: 99%

“…The four major components of an Android application are Activity, Service, Broadcast Receiver, and Content Provider (Wei et al, 2019). Some work (Chen et al, 2014;Nath and Mehtre, 2014;Avdiienko et al, 2015;da Costa et al, 2016;Saracino et al, 2018;Vidas et al, 2014) considered action as features in Malicious application recognition. (Chen et al, 2014) removed the number of exercises and different features to identify malignant applications.…”

Section: App Componentmentioning

confidence: 99%

Malicious Applications Detection in Android Using Machine Learning

Asif¹,

Asif²,

Mubarik³

et al. 2022

IJAIML

1

0

View full text Add to dashboard Cite

A huge number of applications available for Android-based smartphone devices have emerged over the past years. Due to which a huge number of malicious applications has been growing explosively. Many approaches have been proposed to ensure the security and quality of application in the markets. Usually, Machine Learning approaches are utilized in the classification process of malicious application detection. Calculating accurate results of characterizing applications behaviors, or other features, has a direct effect on the results with Machine Learning calculations. Android applications emerge so quickly. The behavior of current applications has gotten progressively malicious. The extraction of malware-infected features from applications is thus become a difficult task. According to our knowledge, a ton of features have been extricated in existing work however no survey has overviewed the features built for identifying malicious applications efficiently. In this paper, we will in general give an extensive review of such sort of work that identifies feature applications by describing various practices of uses with various kinds of features. In this survey we have discussed the following dimensions: extraction and selection of feature methods if any, methods of detection and evaluation performed. In light of our review, we notice the issues of investigating malware-affected features from applications, give the scientific categorization and demonstrate the future headings.

show abstract

“…Their results conclude that there is a promising path with the use of RBMs in learning such characteristics. Another interesting work was proposed by Costa et al [17], which concerns the analysis using unsupervised learning of features in the context of mobile malware detection. This work, however, adopts a new approach, attaching quaternions in the learning process, in order the reach better results.…”

Section: Introductionmentioning

confidence: 99%

Fine‐tuning restricted Boltzmann machines using quaternions and its application for spam detection

Silva

¹

,

Costa

²

,

Papa

³

et al. 2019

View full text Add to dashboard Cite

Restricted Boltzmann Machines (RBMs) have been used in a number of applications, but only a few works have addressed them in the context of information security. However, such models have their performance severely affected by some hyperparameters that are usually hand-tuned. In this work, the authors consider learning features in an unsupervised fashion by means of RBMs fine-tuned by hypercomplex-based metaheuristic techniques in the context of malicious content detection. Experiments are conducted over three public datasets and six metaheuristic techniques, which are used to fine-tune RBM hyperparameters such that RBM extracts features that best represent malicious content present in spam e-mail messages, and generates a dataset to be used as input to classification through the Optimum Path Forest supervised algorithm. Experimental results demonstrate that a small number of features generated through RBM can achieve a competitive accuracy in relation to the original dataset, however, with lower computational cost. Furthermore, this study presents the power of quaternions for RBMs parameter optimisation, comparing it against the well-known Harmonic Search, as well as its variants Improved Harmonic Search and Parameter Setting-Free Harmonic Search. It was concluded that RBM-based learning techniques are suitable for features extraction in the context of this work.

show abstract

Malware Detection in Android-Based Mobile Environments Using Optimum-Path Forest

Cited by 3 publications

References 24 publications

Malicious Applications Detection in Android using Machine Learning

Malicious Applications Detection in Android using Machine Learning

Malicious Applications Detection in Android Using Machine Learning

Fine‐tuning restricted Boltzmann machines using quaternions and its application for spam detection

Contact Info

Product

Resources

About