<p><em>Convolutional Neural Network</em> (CNN) adalah salah satu metode <em>multilayer perceptron</em> yang dapat melakukan klasifikasi aplikasi lebih dari dua kelas. Penelitian ini mengklasifikasikan aplikasi ke dalam tiga kelas, yaitu kelas aplikasi tidak berbahaya, mengandung <em>malware</em> kurang berbahaya, dan mengandung <em>malware</em> berbahaya. Dataset yang digunakan pada penelitian ini terdiri dari <em>dataset</em> Androsec dan Koodous dengan total data 37289 aplikasi. <em>Dataset</em> mengandung aplikasi <em>undetected</em> (tidak mengandung <em>malware</em>) dan <em>detected</em> (mengandung <em>malware</em>). Data <em>detected</em> perlu dikelompokkan dengan algoritme <em>k-means </em>sehingga menghasilkan kelompok aplikasi kurang berbahaya dan berbahaya berdasarkan tingkat kemiripan fitur <em>permission</em> yang dimiliki aplikasi. Kerangka kerja meliputi <em>dataset preprocessing, learning and classification algorithm using CNN</em>, dan <em>check APK to Model</em>. Tingkat akurasi terbaik yang didapat pada penelitian ini adalah 92,23% dan dapat mengklasifikasikan ke dalam kelas tidak berbahaya, kurang berbahaya, dan berbahaya.</p><p> </p><p><em><strong>Abstract</strong></em></p><p class="Judul2"><em>Convolutional Neural Network (CNN) is a multilayer perceptron method which able to classify apps more than two classes. This paper describes classification into three classes such as benign/no malware, less harmful, and harmful application. In this research, we use and construct dataset from Androsec and Koodous with total 37289 apps. Dataset consists of undetected (no malware) and detected (consists of malware). Detected files need to clustered with k-means algorithm to clasify apps into less harmful and harmful </em><em>based on apps permission similarity.</em><em> The framework includes dataset preprocessing, learning and classification algorithm using CNN, and check APK to Model. In this research, we get the best accuracy 92,23% and able to classify apps into three classes benign, less harmful, and harmful.</em><em></em></p><p><em><strong><br /></strong></em></p>
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2025 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.