Research And Implementation Of Android Phishing Hijacking Detection And Protection System

With the rapid development of the mobile Internet and bringing a lot of intangible convenience to people,the problems that endanger people’s information and property security,such as malware attacks and-mobile phone fraud,are becoming more and more serious.Among them,phishing and hijacking has caused great troubles to people’s daily life because of malicious attacks that are not easy to detect and are extremely harmful to steal users’ private data and thereby steal property.The openness and complexity of the Android platform has brought many security problems to itself.Although the Android system has implemented a complete security mechanism as much as possible,ordinary users are easily attacked due to the concealment and unpredictability of phishing hijacking.It causes great harm to users,so it is of great significance for the detection and protection of Android phishing hijacking.This topic uses the concept of container,applies it to Android phishing hijacking protection,and then adds the more popular machine learning technology in recent years,introduces it into the detection of Android phishing hijacking websites,and proposes a new comprehensive detection defense plan.The main work of this paper is as follows:1.A phishing hijacking detection and protection scheme for Android mobile devices is proposed.The scheme includes a secure container protection scheme that can protect users’ applications from phishing and hijacking malicious application attacks on the Android platform,and an Android phishing hijacking scheme that can detect whether the website the user is currently visiting on an Android mobile device is a phishing and hijacking website.Test platform test plan.2.Design and implement a secure container for android phishing and hijacking applications at the application layer level.The security container includes the proxy system service module,the interception system service module and the virtual system service module.This container can form an independent environment isolated from the outside by placing important applications in the container through the HOOK important interface,effectively preventing the attack of phishing and hijacking software.3.Introduce machine learning algorithm to implement an Android phishing hijacking detection platform.Using mainstream binary classification algorithms to train the model,including logistic regression algorithm,decision tree algorithm,random forest algorithm and XGBoost algorithm,it can detect whether the website visited by the user is a phishing and hijacking website.