Android Malware Detection Using Machine Learning Technique

Malware is a common term used to describe different types of malicious activities or inappropriate applications, such as viruses, worms, spyware, Trojans, rootkits and backdoors. The primary characteristic of any malware is its intent to compromise, damage, disrupt, or steal from the targeted Android phone. In a computer environment, typical malware can infect all user programs within the computer operating system, such as applications. However, while research efforts have been directed towards preventing malicious software on personal computers, the same level of attention hasn't been extended to mobile devices, despite the increasing popularity of mobile application development. The aim of this project is to automate the analysis of Android data files using heuristic or machine learning techniques. This study has adopted the Scrum methodology, as it is well-suited for projects in the Machine Learning domain. With Scrum, we can assess the accuracy improvement of the data sets during each sprint, providing an effective means of reviewing the sprint process. The goal is to develop a system capable of identifying new viruses and disseminating that information to all mobile devices, thereby empowering them to defend against future assaults.