An intelligent network user behavior analysis system based on collaborative Markov model and distributed data processing

Tao Wu, Hui He, Xi-Qian Gu, Ying Peng, Yi Zhang, Yuntao Zhou, Senzhe Xu

Abstract: With the boom of internet and ecommerce, increasing numbers of people are getting accustomed to online shopping. Consequently, the analysis of network user behavior is of vital importance to the growth of internet enterprise business. In this paper, an intelligent network user behavior analysis system is illustrated based on collaborative Markov model and distributed data processing. As Markov recommendation algorithm is accurate in prediction, it could be able to solve a series of typical problems like cold start to analyze user behavior data after improving by Slope One algorithm compared with the traditional model. Moreover, distributed computing could divide one problem into multiple tasks and share numerous resources. It could process massive data effectively. Therefore, users would acquire the resources which they truly need among massive data conveniently.

External IDs:dblp:conf/cscwd/WuHGPZZX13