Exploring SVM Kernel and Voting Classifiers for Malware Classification in Android Permission Systems

Sometimes referred to as hostile software, malware is a subset of programming codes aimed to exploit weaknesses in computer system security. Not only through data theft but also by letting Distributed Denial of Service (DDoS) attacks, malware seriously threatens companies. Nowadays, machine learning is a required technique for spotting these threats since the diversity and volume of malware keep increasing. This study aims to determine whether an application device's requested rights are benign or questionable. In performance analysis, the Voting Classifier (VC) ensemble approach was better than the Support Vector Machine (SVM) algorithm with an accuracy rating of 97.9%. Anti-malware preventive measures support the guarantee of safe digital surroundings.