A Novel Web Attack Detection Mechanism Using Maximal-Munch With Torrent Deep Network

A web attack is a harmful and deliberate attempt made by one person or group to gain access to another person's or group's data collection. Due to the incompatibility of the training algorithm for the Cross-Site Scripting (XSS) detection technique and the heterogeneity of attack load, the website was more frequently impacted by the detection of SQL injection attacks. Also, the language of the online sites has a significant impact on how well the current phishing detection system works, which is still a difficult issue. To address these problems, a novel Praise-Worthy Authentication technique is proposed which accurately detects phishing websites by checking the webpage's conformance using the hyperlink property. Also, a Maximal-Munch Algorithm-based ANN is proposed to prevent XSS attacks. The URLs associated with each webpage that is dragged will be sorted out to acquire URL parameters, and text patterns are matched at regular intervals to detect the XSS attack. This work also employs a Torrent Deep network with weight-bolster Algorithm to identify SQL injection by hackers, preventing significant network damage that would otherwise cause data leaks and website paralysis. This proposed Web-strafe Detection Framework has considerably increased the security of websites by identifying numerous threats.