Cơ chế máy học chấn đoán virus máy tính

When computer virus wide spreads in the world nowadays, anti-virus needs to improve their identifying methods to enhance the performance. In this paper, we introduce a new method to diagnose computer virus. First, we analyse the characteristics of viral data type to define virus classes through object-oriented methods. Second, we study the machine learnning mechanism for each virus class. Finally, we apply these learning forms to a data processing stage of a machine learning anti-virus expert system. The experimentation results show that the machine learning approach is suitable for anti-virus to identify the computer virus. This approach also gives a new aspect of anti-virus technology.