Explore figures and images from publications
Frequency Character Distribution of total dataset versus 20% of total dataset discriminated by domain type
Domain Name Service is a central part of Internet regular operation. Such importance has made it a common target of different malicious behaviors such as the application of Domain Generation Algorithms (DGA) for command and control a group of infected computers or Tunneling techniques for bypassing system administrator restrictions. A common detect...
Context in source publication
... are evaluated, the tuning becomes costly in terms of time. To reduce this time, instead of training each model with the complete dataset, only a fraction of the dataset was used. To ensure that this subset is representative of the whole data, we analyzed the Frequency Character Distribution (FCD) of each domain type. As can be seen from Fig. 2, when a fraction of 20% is selected, the FCD of each domain type in the subset is almost exactly the same as if all data were considered. Due to domains that conform the subset are chosen randomly from the complete dataset, the experiment was repeated 30 times and the average and standard deviation are plotted in FCD chart. The optimal ...