Early Warning of Active Worms based on Multi-similarity (VB.NET)
Proceedings of the Fourth International Conference on Machine Learning and Cybernetics, Guangzhou,


Abstract:
Worm detection methods play an important role as frequent breakouts of Internet worm result in tremendous economic destruction. On the basis of analyzing characteristics of normal network traffic distribution, an early worm detection method based on multi-similarity is proposed. It integrates the worms’ behavior attribute with its traffic distribution and detects abnormal behavior by its distribution similarity of its certain features. According to the network simulation experiments, the detection method can find out the worms intrusion against the large-scale network traffic, which does not arouse the sharp changes of the network traffic.