Solution for The CDMC 2017

Abstract

The CDMC 2017 is a competition focusing on real-world problems regarding cybersecurity. We took part in this competition and our team was the first place winner. In this paper, we describe how we solved the following tasks with the provided dataset. We used the Random Forest classifier for all the tasks with the hyperparameter optimization and the feature selection. Experiments showed that our proposed method can obtain an accuracy more than 90% without high computational costs.