The Need for Speed: An Analysis of Brazilian Malware Classifiers
Fabrício Ceschin, Felipe Azevedo Pinage, Marcos Alexandre Castilho, David Menotti, Luiz S. Oliveira, Andre Ricardo Abed Gregio · IEEE Security & Privacy · 2018
Using a dataset containing about 50,000 samples from Brazilian cyberspace, we show that relying solely on conventional machine-learning systems without taking into account the change of the subject's concept decreases the performance of classification, emphasizing the need to update the decision model immediately after concept drift occurs.