A novel parallel classifier scheme for vulnerability detection in Android

摘要: Abstract Android is one of the most commonly used mobile operating systems; however, its open-source nature and flexibility usage attract a lot attention from cybercriminals. In recent years, rapid increase in malware has become major cause concern amongst users. The cybercriminals either aim to exploit confidential information users or try corrupt their systems by infecting them with malicious code. order make more secure, several detection techniques using static, dynamic, hybrid analysis have been introduced times; these are inaccurate low efficiency. paper not only explains how distinctive parallel classifiers can be for detecting zero-day android but also addresses oncoming highly elusive vulnerabilities. proposed methodology combines characteristics various expectation maximization achieve 98.27% accuracy.