Researchers have developed a new machine learning approach for operating system fingerprinting using the CIC-IDS2017 dataset. The proposed tool, OsirisML, preprocesses network traffic data with nPrint and then applies XGBoost to train and test machine learning models. This method achieved a high accuracy of 97.66% on a subset of the Friday capture and 84.69% on the full Friday capture when data was randomly split for training and testing. For the Monday capture, OsirisML reached 73.83% accuracy and a 79.38% F-1 score. AI
IMPACT This research could improve network security by enabling more accurate identification of operating systems, potentially aiding in threat detection and network management.
RANK_REASON Academic paper detailing a new machine learning method and tool for OS fingerprinting. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
