一种基于模糊模式的脚本病毒检测方法

In this paper, a method of classification based on fuzzy pattern model is designed to detect script viruses. Firstly, extract keywords matrix of script samples from the virus script collection and the benign script collection; secondly, select normal distribution of partial large-scale to construct the membership function according to the script virus features; then, create benign script pattern and virus script pattern by using samples' keywords matrix and the selected membership function. In order to design a proper fuzzy pattern model, we conduct a detailed analysis on Euclidean close degree distribution and the selection of membership function. In the end, the test results of the designed fuzzy pattern model are displayed..