<ul><li>A new machine learning model has been introduced for generating synthetic network flow datasets that closely resemble real-world networks.</li><li>The approach involves creating dynamic multigraphs using a stochastic Kronecker graph generator for structure generation and a tabular generative adversarial network for feature generation.</li><li>An XGBoost model is employed for graph alignment to ensure accurate overlay of features onto the generated graph structure.</li><li>The model demonstrates improved accuracy over previous large-scale graph generation methods while maintaining efficiency and explores the trade-off between accuracy and diversity in synthetic graph dataset creation.</li></ul>

Synthesizing Diverse Network Flow Datasets with Scalable Dynamic Multigraph Generation

Discover more