Ensemble-based Feature Selection and Classification Model for DNS Typo-squatting Detection

摘要: Domain Name System (DNS) plays in important role the current IP-based Internet architecture. This is because it performs domain name to IP resolution. However, DNS protocol has several security vulnerabilities due lack of data integrity and origin authentication within it. paper focuses on one particular vulnerability, namely typo-squatting. Typo-squatting refers registration a that extremely similar an existing popular brand with goal redirecting users malicious/suspicious websites. The danger typo-squatting can lead information threat, corporate secret leakage, facilitate fraud. builds our previous work [1], which only proposed majority-voting based classifier, by proposing ensemble-based feature selection bagging classification model detect attack. Experimental results show framework achieves high accuracy precision identifying domains (a loss at most 1.5% 5% when compared used complete set) while having lower computational complexity smaller set reduction more than 50% size).