Method and device for identifying XSS attack, and computer readable storage medium

Abstract

The embodiment of the invention discloses a method and device for identifying an XSS attack, and a computer-readable storage medium. The data to be detected is preprocessed to obtain a digital vector. the trained character-based convolutional neural network model is used to process the digital vector to obtain the attack probability of XSS. When the XSS attack probability value is greater than orequal to the preset threshold, the output data to be detected is XSS attack data. Compared with the prior art, the technical scheme does not need to make rules artificially, and only needs to use thetrained character-type convolution neural network model for XSS attack analysis. The model uses the network structure of convolution neural network, and can identify some attacks with similar structure, which improves the generalization ability of the model. And the character-based convolution neural network model can take each character as the smallest processing unit. Compared with word segmentation processing, it can learn deeper semantics and improve the accuracy of XSS attack recognition.

Description

technical field [0001] The present invention relates to the technical field of data protection, in particular to a method, device and computer-readable storage medium for identifying XSS attacks. Background technique [0002] Cross Site Scripting (Cross Site Scripting, XSS) is a kind of computer security vulnerability in web applications. Malicious attackers insert malicious script codes into web pages. When users browse the page, the script codes embedded in the web will be destroyed. Execution, so as to achieve the purpose of maliciously attacking users. [0003] The current mainstream identification method is to use manually written regular rules to match suspicious structures in input information, such as <img SRC＝"javascript：alert('XSS')；"> Can be hit by the regular rule alert(.*). [0004] However, this method relies heavily on the experience of rule makers, and is prone to omissions and misjudgments. And once a variant attack occurs, the existing rules may becom...

AI Technical Summary

Problems solved by technology

[0004] However, this method relies heavily on the experience of rule makers, and is prone to omissions and misjudgments.

And once a variant attack occurs, the existing rules may become invalid, and new rules need to be formulated to identify the attack

Images