Mining And Application of User Behavior Pattern Based on Operation And Maintenance Data

In order to provide users with personalized services, or to implement user-centric management, operators usually need to collect various information of users. However, in an actual network scenario, such information is often difficult to obtain accurately due to reasons such as protecting user privacy and avoiding interference with the user experience process. Common methods such as setting up a laboratory network environment, implementing user research, etc., are difficult to implement in the live network. Through the service log of the network device interface, this paper extracts a variety of data related to the user’s network behavior, and proposes a comprehensive multi-dimensional data user behavior expression method, which is transformed into the expression of user behavior, including time, space and behavior semantics. For the user’s daily network behavior, we propose a user behavior symbolization method for different application scenarios, and propose a user behavior pattern mining method based on PrefixSpan to mine the user behavior sequence pattern after symbolization. This method can mine the patterns of user behavior and behavior from the data, and provide the basis for personalized service and management. We used the http connection log of all mobile users of a carrier in a city for one day to conduct our experiments.

[1]  Wen-Chih Peng,et al.  Mining Spatial-Temporal Semantic Trajectory Patterns from Raw Trajectories , 2015, 2015 IEEE International Conference on Data Mining Workshop (ICDMW).

[2]  M. A. Dorairangaswamy,et al.  Discovering User Pattern Analysis from Web Log Data using Weblog Expert , 2016 .

[3]  Janak Thakkar,et al.  A Survey On Efficient Frequent Pattern Mining Techniques , 2018 .

[4]  Qiming Chen,et al.  PrefixSpan,: mining sequential patterns efficiently by prefix-projected pattern growth , 2001, Proceedings 17th International Conference on Data Engineering.

[5]  Lidan Shou,et al.  Splitter: Mining Fine-Grained Sequential Patterns in Semantic Trajectories , 2014, Proc. VLDB Endow..

[6]  Kai Zheng,et al.  STMaker - A System to Make Sense of Trajectory Data , 2014, Proc. VLDB Endow..

[7]  John Keeney,et al.  Multilevel pattern mining architecture for automatic network monitoring in heterogeneous wireless communication networks , 2016, China Communications.

[8]  Yun Sing Koh,et al.  A Survey of Sequential Pattern Mining , 2017 .

[9]  C. Siva Ram Murthy,et al.  A learning based mobile user traffic characterization for efficient resource management in cellular networks , 2015, 2015 12th Annual IEEE Consumer Communications and Networking Conference (CCNC).

[10]  Om Prakash.  Analyzing and Predicting User Behavior Pattern from Weblogs , 2018 .

[11]  Enhong Chen,et al.  Mining Mobile User Preferences for Personalized Context-Aware Recommendation , 2014, ACM Trans. Intell. Syst. Technol..