Research And Implementation Of Detecting System For Malicious WEB Pages

The Internet has almost become an indispensable part of modern life.It has greatly enriched people's lives and also greatly facilitated business.Now whether it is business units or government agencies to establish their own website has become a trend.This trend is in line with the development of the times.Many of the businesses that preceded them were handled offline,leaving the scene always overcrowded and the state needing a lot of manpower to handle it,and even then it was just a drop in the bucket,and the order and fire problems caused by the large numbers of people gathered were even more difficult to tell.After the business is networked,taking banks and train stations as an example,people can now make online transfers and self-service ticketing,greatly reducing on-site pressure.Only a handful of businesses need to be handled on site,reducing the burden on both parties.However,with the rapid development of online services,various attacks against the Internet have also become more and more frequent.Users' property safety and privacy security face enormous risks.Various malicious web pages emerge in endless streams,with losses of up to billions each year.Therefore,the importance of malicious web page detection is slowly emerging,and enhancing malicious web page detection capabilities is of great practical significance to both users and website owners.The main work of this paper includes the following aspects:1.Aiming at the weakness of existing dynamic detection module based on virtual machine and its low timeliness,a hybrid malicious web page detection method based on simulated browser and classification algorithm is proposed.Combined with the advantages of the two detection techniques,the detection can be completed quickly while ensuring the detection rate of malicious web pages.2.In this paper,the commonly used means of attack on malicious web pages are researched.The research on url obfuscation attack and XSS vulnerability detection is carried out.The dynamic stain analysis algorithm is improved to detect the existence of XSS vulnerability in the web site.In view of the shortcomings of simply matching url in blacklist,List algorithm that identifies URLs that contain similar malicious content,and ultimately implements dynamic detection modules based on black-and-white lists and simulated browsers.3.Investigate the commonly used classification algorithms,analyze the features of malicious web pages and implement feature extraction functions,and use the test data to analyze the performance of several analysis algorithms.Finally,SVM is used as a classifier to implement a static detection module based on classification algorithm.4.The final investigation of the network development technology,and for user login encryption to prevent theft of cookie attacks to improve service security.Finally,the entire system of network services,enabling users to log in through an online testing platform anytime,anywhere and the use of detection capabilities,no longer need to repeat the deployment of services,of practical significance.