作者: Pratiksha Natani , Deepti Vidyarthi
DOI: 10.1007/978-3-642-40576-1_37
关键词: Classifier (UML) 、 Application programming interface 、 Artificial intelligence 、 Computer science 、 Directory 、 Boosting (machine learning) 、 Upload 、 Unavailability 、 Malware 、 Feature vector 、 Data mining 、 Machine learning
摘要: Malicious code, known as malware, when executed can steal information, damage the system or may cause unavailability of resources. In order to safeguard information systems from effective detection malware is a top priority task. Malware exhibits malicious behaviors like connecting remote host, downloading file creating in directory etc. These be mapped functions used by files which are imported system’s dynamic link libraries i.e. Application programming interface (API) functions. Hence, we propose technique detect using API function frequency feature vector for classifying file. We use Ensemble based classifier classification, it proven stable and robust classification technique. Experiments conducted over 200 classified effectively. Bagging ensemble provides better results compared boosting. Comparison with other techniques also listed.