Research On Phishing Website Detection Based On Cloud Computing

Phishing is an attack wherein the attacker exploits social engineering techniques to perform identity theft.The URL,content and layout of phishing websites are very similar to those of real websites,Netizens without security awareness are easily deceived.With the development of networks and electronic commerce,more and more people are carrying out economic activities through the Internet,Since the economic losses and social problems caused by phishing websites have increased year by year,research on phishing websites' detection technologies is of utmost importance.For phishing detection problems,researchers have put forward many identification technologies and solutions,However,attackers are also constantly adopting some new technical means to enhance the concealment of phishing websites in order to increase attack efficiency,which has led to a single phishing website detection technology that is difficult to deal with increasingly sophisticated fishing techniques;At the same time,the survival time of phishing websites is short,and the traditional phishing website detection technology is generally lack of real-time.Therefore,this article launches a research on the detection of phishing websites.The main contents of this paper are as follows:(1)An improved Trust Rank algorithm is proposed to detect phishing websites.The Trust Rank algorithm can effectively detect spam pages,and phishing websites also belong to spam pages.Similarly,phishing websites also belong to spam pages,and phishing websites and spam pages have similar characteristics in relation to links.The Trust Rank algorithm is used to improve the detection of phishing websites by setting up promotion page collections and combining phishing site features,and a parallel processing scheme for the algorithms is designed under the cloud computing environment to increase the data processing rate and achieve the purpose of real-time detection.(2)A phishing website detection algorithm based on link structure is proposed.The phishing websites are not real websites.There are fixed differences between the two different natures of the websites in the link structure.By analyzing the link structure relationships of the websites and extracting relevant features,the K-means algorithm is used to quantify the link structure relationships of the websites,At the same time,the K-means algorithm's parallelization processing scheme is designed in the cloud computing environment to improve the detection rate of the algorithm.(3)A phishing website collaborative detection model based on approaches degree is proposed.The relationship of detection results between phishing website detection algorithm based on improved Trust Rank and phishing website detection algorithm based on link structure is analyzed.Based on this,a phishing website collaborative detection model based on approaches degree is proposed,and through experimental comparisons to verify the advantages of the model,At the same time,the cloud computing processing scheme is introduced to improve the data processing capability and detection rate.