<![CDATA[Not a few people abuse technology to commit crimes or what is called a cybercrime. One form of cybercrime is a malware or malicious software attacks. It is necessary to detect malware attacks so that users can find out whether the data comes from the internet or whether an Android application is safe from inserting malware or not. Malware is quite difficult to classify and differentiate directly; therefore, we need a way to classify good websites and malicious websites. This study uses the K-Nearest Neighbor method to classify android application malware. This study uses Android Malware/Benign Permissions data in the form of CSV files obtained from Kaggle.com. The results showed that malware classification and not malware in the android application permit could be done well using the K-Nearest Neighbor algorithm, which produces an accuracy of 77%. Classification of malware and not malware is better done by combining the K-Nearest Neighbor method with feature selection, which increases the accuracy value from 44% to 77%.]]>
Copyrights © 2023
