基于Web页面特征的反钓鱼系统的设计与实现

发布时间：2018-05-26 07:51

本文选题：反钓鱼 + URL特征　；参考：《北京邮电大学》2014年硕士论文

【摘要】：当今的互联网环境中,网络木马与病毒肆虐的同时,也充斥着大量的钓鱼网站。网络钓鱼是一种网络欺诈手段,攻击者精心设计一个与目标网站十分相似的钓鱼网站,或具有虚假信息的网站,一旦受害者访问该网站并轻信网站的内容,攻击者就可能从中获取受害者的敏感信息,如账号、密码等,或者造成直接财产损失。网络钓鱼是一种利用社会工程学手段的攻击方式,钓鱼网站的制作不需要太多的技术含量,而是利用人的心理弱点进行欺骗,因此钓鱼网站曾经是最易被忽视的网络安全问题。网络钓鱼给互联网安全造成了重大威胁,严重损害了网络用户的利益。刚刚结束不久的2013年“双11”各大电商创造了网络交易的新神话,同时也给网络钓鱼带来了欺诈的好时机,根据中国反钓鱼网站联盟钓鱼网站处理简报的数据,2013年10月联盟处理的钓鱼网站数量近同年9月的两倍,并且支付交易类,金融证券类等可以直接给网络用户造成财产损失的钓鱼网站的总数,占钓鱼网站总量的大半,因此打击钓鱼网站是互联网各界义不容辞的责任。本文的目的是设计并实现反钓鱼系统,下面对本文的研究内容以及主要工作进行归纳： 1、本文综述了目前主要的反网络钓鱼技术,包括基于黑白名单的网络钓鱼检测机制,基于页面内容的启发式网络钓鱼检测机制,基于视觉相似的网络钓鱼检测机制,并且对上述三种反钓鱼技术的优缺点进行了归纳总结。本文根据各大互联网安全报告、中国反钓鱼网站联盟提供的数据等资料以及通过对互联网实时检测的钓鱼网站结果的长期研究,总结出了钓鱼网站存在的四个趋势。 2、本文结合现有的反钓鱼技术以及钓鱼网站存在的趋势,对反钓鱼系统进行了概要设计,对不同特点的钓鱼网站采用不同的检测方式。本文根据钓鱼网站高度模仿目标网站的特性,引入了网页分类技术对这部分钓鱼网站进行重点检测。由于网络钓鱼的制作已经形成了完整的产业链,钓鱼网站存在批量生产的特性,本文据此特性引入了网页去重技术,利用已经获取的钓鱼网站检测与之相似的钓鱼网站。 3、本文通过挖掘钓鱼网站URL的特征以及页面内容的特征对反钓鱼系统进行了编码实现。
[Abstract]:In today's Internet environment, the network Trojan and virus rampant, but also a large number of fishing sites. Phishing is a form of cyberfraud in which an attacker designs a phishing site that is very similar to the target site, or a website with false information, once the victim visits the site and believes its content. The attacker may obtain sensitive information about the victim, such as account number, password, etc., or cause direct property damage. Phishing is an attack using social engineering means. Phishing websites do not need too much technology content, but make use of human psychological weakness to cheat, so phishing website was once the most neglected network security problem. Phishing poses a serious threat to Internet security and seriously damages the interests of Internet users. The recent completion of the 2013 "double 11" ecommerce has created a new myth of online trading, and has also given phishing a good time to cheat. In October 2013, the alliance handled nearly twice as many phishing sites as it did in September of the same year, and paid for transactions, according to the China Anti-phishing website Coalition phishing website processing briefing. The total number of phishing websites which can directly cause property losses to network users accounts for most of the total phishing websites, so it is incumbent on all circles of the Internet to crack down on phishing websites. The purpose of this paper is to design and implement the anti-phishing system. 1. This paper summarizes the main anti-phishing technologies, including phishing detection mechanism based on black-and-white list, heuristic phishing detection mechanism based on page content, phishing detection mechanism based on visual similarity. The advantages and disadvantages of the above three anti-fishing techniques are summarized. Based on the reports of Internet security, the data provided by China Anti-phishing website Association and the long-term research on the results of phishing websites detected in real time by the Internet, four trends of phishing websites are summarized in this paper. 2. Combined with the existing anti-phishing technology and the existing trend of fishing website, this paper gives a brief design of anti-phishing system, and adopts different detection methods for different fishing websites with different characteristics. According to the characteristics of the target website, this paper introduces the technology of web page classification to detect the phishing website. As the production of phishing has formed a complete industrial chain and the phishing website has the characteristics of mass production, this paper introduces the technology of webpage de-reduplication, and uses the obtained phishing website to detect similar phishing sites. 3. In this paper, the anti-phishing system is coded by mining the features of URL and page content.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP393.08

【参考文献】