DNS (Domain Name System) is the distributed computer service that associates Internet domain names with their IP addresses. For the sake of business or malicious actions, some persons can make various DNS abuses such as squatting, typosquatting, and so on. To prevent such abuses, DNS managers elaborate a couple of policies and tools to double-check whether a domain name is compliant or not. Some existing solutions rely on identifying a list of reserved terms and proceed to syntactic verification before allowing the record of a new domain name. Such an approach, unfortunately, does not prevent typosquatting and Soundsquatting. To overcome such a drawback, we introduce a control approach made of both syntactic and phonetic verification supported by a classification module for decision-making. Our approach is validated over a set of 9726 domain names and around 5200 reserved terms. Results show the effectiveness of our approach and how it overcomes the existing algorithms devised for terms-reserved compliance check.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

AI-Based Control Approach of .SN Reserved Domain Names (aIDN.SN)

  • Evrard Cabrel Nguemeyou Tchouangang,
  • Ahmadou Ndiaye,
  • Bassirou Kassé,
  • Alex Corenthin,
  • Idrissa Sarr

摘要

DNS (Domain Name System) is the distributed computer service that associates Internet domain names with their IP addresses. For the sake of business or malicious actions, some persons can make various DNS abuses such as squatting, typosquatting, and so on. To prevent such abuses, DNS managers elaborate a couple of policies and tools to double-check whether a domain name is compliant or not. Some existing solutions rely on identifying a list of reserved terms and proceed to syntactic verification before allowing the record of a new domain name. Such an approach, unfortunately, does not prevent typosquatting and Soundsquatting. To overcome such a drawback, we introduce a control approach made of both syntactic and phonetic verification supported by a classification module for decision-making. Our approach is validated over a set of 9726 domain names and around 5200 reserved terms. Results show the effectiveness of our approach and how it overcomes the existing algorithms devised for terms-reserved compliance check.