In the process of malicious code detection

it is assumed that the malicious code hidden deep

it is difficult to complete and accurate of malicious code features are extracted. Malicious code detection using the traditional algorithm

the distribution of malicious code is unknown

without fully considering the characteristic differences between different categories code

reduces the accuracy of the malicious code detection. To this end

the malicious code detection based on fuzzy recognition method is proposed. According to the theory of support vector machine (SVM) related to extract characteristics of malicious code

and will be the basis of the above characteristics as malicious code identification. Establish a fuzzy recognition to identify tree

calculated to identify objects belong to the probability of malicious code

for detecting malicious code. The experimental results show that the improved algorithm of malicious code detection

can greatly improve the accuracy of detection.