Abstract:

Basing on the researching of the relation between personal and the culture he is in and the interactive between he and the other persons，we can find a lot of character of the person by analysis of egocentric network.Recently，data mining is more and more widely used in the telecommunication area，and egocentric is a new idea that trade the person as a part of the whole network.But because of the size of data，dimensionality of data and the complication of the computation，traditional methods is not suitable for this kind of application.In this article，we give the implementation of the traditional egocentric algorithm based on mapreduce module.And then we propose a new egocentric network generating algorithm based on the discovery of triangles.After that，we give the implementation of the new algorithm based on the mapreduce module.In this new algorithm，we make some optimize in connection with the mapreduce module and the character of the real social network to enhance the efficiency.At last，we will compare these two algorithms by time consumed and IO.

Key words: social network analysis, distribute, hadoop, mapreduce, data mining, egocentric network