Towards quantifying visual similarity of domain names for combating typosquatting abuse

摘要: Typosquatting becomes a speculative and serious phenomenon for both Internet users brand owners of popular websites. Typosquatters register similar domain names websites to profit from displaying advertisements, redirecting traffic third-party pages, deploying phishing sites, or serving malware. Thus, much work have been done on measuring typosquatting in distribution, monetization cost etc. This paper does not measure typosquatting, but tries combat abuse the abnormal detection view: that looks very like one website is suspicious. We propose TypoPegging, reverse lookup approach quickly accurately get most given domain. Specifically, we novel quantitative method visual similarity two domains. The proposed based generalized Levenshtein distance takes insights our characteristics. Then give an efficient search maximum over set. accelerate searching process triangle inequality metric locality sensitive hashing algorithm. Preliminary results show effective differentiating normal ones. can also speedup orders magnitude comparing with linear method.