Internet usage modeling of large wireless networks using self-organizing maps

摘要

Human behavior and interests will play a central role in future mobile networks. We introduce a systematic method for large-scale multi-dimensional analysis of online activity for thousands of mobile users over 100 web domains. We propose a modeling approach based on self-organizing maps for discovering, organizing and visualizing different mobile users' trends from billions of WLAN and netflow records. We find surprisingly that users' trends can be modeled using a self-organizing map with clearly distinct characteristics while similar domains are modeled by neighboring nodes. This is the first study to acquire such detailed results for mobile Internet usage.