Next, we study the data collection and data processing steps of the commontraffic analysis pipeline. NPrintML, a generic feature engineering and model training solution. Integrate nPrint with automated machine learning to produce Representation learning and model training for a variety of tasks. We develop nPrint, a unified packet representation that is amenable to Training-two of the most manual and painstaking steps for any traffic analysis task. First, we examine the process of feature engineering and model Several stages of the common pipeline can be better accomplished using generic This dissertation questions the assumption that each stage in the shared methodological pipeline should be custom-built to each task, exploring if Methodological pipeline shared across tasks, each instantiated pipeline isĬustom-built for the task at hand, requiring new traffic processing systems, flows, applications), engineer featuresįor the task, and finally train models using the engineered features. Pipeline: curate a network traffic dataset, create a system to separate andĪssociate labels with the traffic (e.g. Machine-learning based traffic analysis techniques shares a common methodological As such, newly developed methodsĪlmost universally leverage machine techniques. Signature-based solutions less effective. Researchers and practitioners rely on network traffic analysis techniquesfor a variety of critical network security and network management tasks.Įver-increasing traffic volumes and encryption rates have rendered traditional, Please use this identifier to cite or link to this item:Ī Generic Framework For Network Traffic Analysis Princeton University Undergraduate Senior Theses, 1924-2022 Princeton University Masters Theses, 2022-2023 Princeton University Doctoral Dissertations, 2011-2023 Princeton School of Public and International Affairs
Liechtenstein Institute on Self-Determination Lewis-Sigler Institute for Integrative Genomics Department of Slavic Languages and Literatures
0 kommentar(er)
