Hadoop-Based Analysis for Large-Scale Click-Through Patterns in 4G Network

摘要

In this work, we focus on understanding what happens under the hood of HTTP traffic. We proposed a systematic approach to reconstruct users click-through patterns in web starting from raw network data. We validated the proposed method using a synthesized network traffic data set and a real-world 4G network traffic data set. The precision calculated by the proposed model is 93.7% and the recall is 91.6%. The experiment results on the real-world 4G network traffic data show that the click-through stream of 4G network is more deeper than that of 3G network.