Analysis of Anomaly detection of Malware using KNN

Computer malware development has grown rapidly in the last decade. Nowadays, malicious software (malware) is widely used by cybercriminals to attack computer systems. Malware detection techniques are most effective when they extract discriminative features from the malware, various static and dynamic tools can be used to set up analysis environments. Using traditional methods to classify malware may have worked in the past, but using machine learning algorithms may be more effective in the future as they are designed to keep up with the complexity and speed of malware development. A comprehensive study of anomaly detection of malware based on machine learning algorithms is presented here. This paper also explains about the implementation of k-nearest neighbors of anomaly detection and discusses the challenges associated with implementing malware classifiers. In the final section, we discuss future directives regarding developing an effective malware detection system.