Android Malware Detection Using Machine Learning with Feature Selection Based on the Genetic Algorithm

Lee, Jaehyeong; Jang, Hyuk Ju; Ha, Sangyul; Yoon, Yourim

doi:10.3390/math9212813

Cited by 33 publications

(26 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Genetic algorithm (GA) is a search and optimization technology inspired by the biological evolution process [ 23 , 24 ]. Based on the rule of survival of the fittest, it searches for the optimal solution through various genetic operations.…”

Section: Mfdroidmentioning

confidence: 99%

MFDroid: A Stacking Ensemble Learning Framework for Android Malware Detection

Wang

Zhang

Zhao

et al. 2022

Sensors

View full text Add to dashboard Cite

As Android is a popular a mobile operating system, Android malware is on the rise, which poses a great threat to user privacy and security. Considering the poor detection effects of the single feature selection algorithm and the low detection efficiency of traditional machine learning methods, we propose an Android malware detection framework based on stacking ensemble learning—MFDroid—to identify Android malware. In this paper, we used seven feature selection algorithms to select permissions, API calls, and opcodes, and then merged the results of each feature selection algorithm to obtain a new feature set. Subsequently, we used this to train the base learner, and set the logical regression as a meta-classifier, to learn the implicit information from the output of base learners and obtain the classification results. After the evaluation, the F1-score of MFDroid reached 96.0%. Finally, we analyzed each type of feature to identify the differences between malicious and benign applications. At the end of this paper, we present some general conclusions. In recent years, malicious applications and benign applications have been similar in terms of permission requests. In other words, the model of training, only with permission, can no longer effectively or efficiently distinguish malicious applications from benign applications.

show abstract

Section: Mfdroidmentioning

confidence: 99%

MFDroid: A Stacking Ensemble Learning Framework for Android Malware Detection

Wang

Zhang

Zhao

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…The papers [29]- [31], [37], [44], use both permissions and API calls, and pass them to different machine-learning algorithms such as random forest, SVM, and logistic regression. By utilizing different datasets, various accuracies were obtained in the range 87 -98%.…”

Section: A Static Analysis Based Malware Detectionmentioning

confidence: 99%

“…The latter proposed an ondevice malware detection and remover, but its accuracy is not given. The study [37] uses permissions and API calls and tries genetic algorithm in the feature selection process. After that, the features are passed to different ML algorithms such as J48, decision tree, random forest, and naive bayes.…”

Section: A Static Analysis Based Malware Detectionmentioning

confidence: 99%

A Method for Automatic Android Malware Detection Based on Static Analysis and Deep Learning

2022

View full text Add to dashboard Cite

The computers nowadays are being replaced by the smartphones for the most of the internet users around the world, and Android is getting the most of the smartphone systems' market. This rise of the usage of smartphones generally, and the Android system specifically, leads to a strong need to effectively secure Android, as the malware developers are targeting it with sophisticated and obfuscated malware applications. Consequently, a lot of studies were performed to propose a robust method to detect and classify android malicious software (malware). Some of them were effective, some were not; with accuracy below 90%, and some of them are being outdated; using datasets that became old containing applications for old versions of Android that are rarely used today. In this paper, a new method is proposed by using static analysis and gathering as most useful features of android applications as possible, along with two new proposed features, and then passing them to a functional API deep learning model we made. This method was implemented on a new and classified android application dataset, using 14079 malware and benign samples in total, with malware samples classified into four malware classes. Two major experiments with this dataset were implemented, one for malware detection with the dataset samples categorized into two classes as just malware and benign, the second one was made for malware detection and classification, using all the five classes of the dataset. As a result, our model overcomes the related works when using just two classes with F1-score of 99.5%. Also, high malware detection and classification performance was obtained by using the five classes, with F1-score of 97%.

show abstract

“…To address the above issues, various works are proposed to solve feature selection problems using metaheuristics [35]. Most of them use genetic algorithms (GA) [36][37][38][39]. Meta-heuristic algorithms based on swarm intelligence are also applied to feature selection, such as ant colony optimization (ACO) [40,41], particle swarm optimization (PSO) [42,43], and bee swarm optimization (BSO) [44,45].…”

Section: Introductionmentioning

confidence: 99%

Framework for feature selection of predicting the diagnosis and prognosis of necrotizing enterocolitis

Song

Yao

et al. 2022

PLoS ONE

View full text Add to dashboard Cite

Neonatal necrotizing enterocolitis (NEC) occurs worldwide and is a major source of neonatal morbidity and mortality. Researchers have developed many methods for predicting NEC diagnosis and prognosis. However, most people use statistical methods to select features, which may ignore the correlation between features. In addition, because they consider a small dimension of characteristics, they neglect some laboratory parameters such as white blood cell count, lymphocyte percentage, and mean platelet volume, which could be potentially influential factors affecting the diagnosis and prognosis of NEC. To address these issues, we include more perinatal, clinical, and laboratory information, including anemia—red blood cell transfusion and feeding strategies, and propose a ridge regression and Q-learning strategy based bee swarm optimization (RQBSO) metaheuristic algorithm for predicting NEC diagnosis and prognosis. Finally, a linear support vector machine (linear SVM), which specializes in classifying high-dimensional features, is used as a classifier. In the NEC diagnostic prediction experiment, the area under the receiver operating characteristic curve (AUROC) of dataset 1 (feeding intolerance + NEC) reaches 94.23%. In the NEC prognostic prediction experiment, the AUROC of dataset 2 (medical NEC + surgical NEC) reaches 91.88%. Additionally, the classification accuracy of the RQBSO algorithm on the NEC dataset is higher than the other feature selection algorithms. Thus, the proposed approach has the potential to identify predictors that contribute to the diagnosis of NEC and stratification of disease severity in a clinical setting.

show abstract

Android Malware Detection Using Machine Learning with Feature Selection Based on the Genetic Algorithm

Cited by 33 publications

References 35 publications

MFDroid: A Stacking Ensemble Learning Framework for Android Malware Detection

MFDroid: A Stacking Ensemble Learning Framework for Android Malware Detection

A Method for Automatic Android Malware Detection Based on Static Analysis and Deep Learning

Framework for feature selection of predicting the diagnosis and prognosis of necrotizing enterocolitis

Contact Info

Product

Resources

About