A hybrid of CNN and LSTM methods of XSS Attacks

Abstract — Cross-site scripting is a common attack on online applications that is vulnerable at the application layer. The attacker injects the script and handles malicious scripts for trusted websites. There are various sorts of XSS scripts that can be used to target extremely open websites and online apps. The attacker could load or redirect the malicious web page.  We then train and test word vectors using CNNs to create a model that can be used in a web application. The performance of the CNN algorithms in identifying XSS vulnerabilities in online applications and web pages was used to train machine learning algorithms. To detect the accuracy of the machine learning algorithm of the model, convolutional neural networks are used.