The spark-based framework for mobile network data and cluster analysis on mobile users´ behaviors

Mining data from mobile network data produced by mobile users becomes a hot issue in recent years. For the growth of these data, how to process and analyze these data becomes a challenging problem. A Spark-based framework is proposed in this paper, which process and analyze real mobile network data. CURE algorithm is implemented in this framework to cluster mobile users´ behaviors parallelly, and discover their preferences in mobile network.