Malware Classification through HEX Conversion and Mining

摘要: The malicious codes are normally referred as malware. Systems vulnerable to the traditional attacks, and attackers continue find new ways around existing protection mechanisms in order execute their injected code. Malware is a pervasive problem distributed computer network systems. These executables created at rate of thousands every year. There several types threat violate these components; for example Viruses, Worms, Trojan horse Malware. represents serious confidentiality since it may result loss control over private data users. It typically hidden from user difficult detect can create significant unwanted CPU activity, disk usage traffic. In systems, programs be detected by automatic signature generation called F-Sign extraction unique signatures malware files. This primarily intended high-speed process based on comparison with common function repository. mining framework employed this research learns through analyzing behavior benign large datasets. We have robust classifiers, namely Naive Bayes (NB) Algorithm, k−Nearest Neighbor (kNN) J48 decision tree evaluated performance. involves extracting opcode sequence dataset, construct classification model identify or benign. Our approach showed 98.4% detection whose was not used building process.