Machine Learning-Based Hybrid Technique for Securing Edge Computing

Mobile edge computing relies heavily on smartphones (MEC). The proper operation of cloud services depends on the security of data stored on mobile devices. Malicious Android applications are becoming more common. To create effective malware detection systems, it is imperative that these apps be thoroughly examined for their malign intent. Machine learning (ML) techniques based on hybrid features may be critical in the detection of Android malware, according to the current most up-to-date model. The feature selection process is critical for capturing the right behavioral patterns of malware instances in order to create a usable categorization of mobile apps. Using both static and dynamic elements of Android applications, we've developed a new method for identifying malware. Malware detection techniques are evaluated on their ability to identify hybrid elements. An F-measure score of 97 percent indicates that the recommended set of features is successful in identifying malware risks. Keywords: Edge computing, smartphone, machine learning, malware detection.