Exploiting Statistical and Structural Features for the Detection of Domain Generation Algorithms

<p>This repository contains a&nbsp;dataset for the research of domain generation algorithms (DGAs) and machine learning. More precisely, it targets dictionary-based DGAs.</p> <p><em>Constantinos Patsakis, Fran Casino: &quot;Exploiting Statistical and Structural Features for the Detection of Domain Generation Algorithms&quot;,&nbsp;Journal of Information Security and Applications, 2021.</em></p> <p>Features ordered as in the shared dataset:</p> <ul> <li>Family: DGA that the domain belongs to</li> <li>SLD: SLD of the Domain</li> <li>L-LEN: The length of Domain</li> <li>L-DIG: The number of digits in Domain</li> <li>L-CON-MAX: The maximum number of consecutive consonants Domain</li> <li>R-CON-VOW: Number of consonants divided by L-LEN&nbsp;</li> <li>L-SYM: The number of special characters</li> <li>R-SYM-LEN: L-SYM divided by L-LEN</li> <li>R-Dom-3G: Ratio of benign grams in Dom-3G</li> <li>R-Dom-4G: Ratio of benign grams in Dom-4G</li> <li>R-Dom-5G: Ratio of benign grams in Dom-5G</li> <li>L-W2: Number of words with more than 2 characters in Domain</li> <li>L-W3: Number of words with more than 3 characters in Domain</li> <li>R-WS-LEN: Dom-WS divided by L-LEN</li> <li>R-WDS-LEN: Dom-WDS divided by L-LEN</li> <li>R-W2-LEN: Dom-W2 divided by L-LEN</li> <li>R-W3-LEN: Dom-W3 divided by L-LEN</li> <li>M2-Dom-Ws: 2-Chain Markov English grams applied to Dom-WS</li> <li>M2-Dom-WDS: 2-Chain Markov English grams applied Dom-WDS</li> <li>E-Dom-WS: Entropy of Dom-WS&nbsp;</li> <li>E-Dom-WDS: Entropy of Dom-WDS</li> <li>E-Dom-W2: Entropy of Dom-W2</li> <li>E-Dom-W3: Entropy of Dom-W3</li> </ul>