Cache Pollution Detection Method Based on GBDT in Information-Centric Network

There is a new cache pollution attack in the information-centric network (ICN), which fills router by sending large number of requests for nonpopular content. This will severely reduce hit rate. Therefore, detection attacks also an urgent problem current information center network. In existing research on detection, most methods manually setting threshold are used detection. The accuracy result depends setting, and adaptability to different environments weak. order improve environments, this paper proposes algorithm based gradient boost decision tree (GBDT), can obtain through model learning. Method. feature selection, uses two features node status path as input, improves method. proves improvement method comparative experiments.