A Fast Identification Approach to Social Network Traffic Based on Unsupervised Learning
Qiu Chen-x · Shuxue de shijian yu renshi · 2014
We combine existing methods of network traffic identification and the features of social network flow after analyzing the development and research statement of social network.We proposed an unsupervised social network flow identification method based on KMeans clustering algorithm.In order to improve the efficiency of processing,real-time,we use the MapReduce model which is provided by the open source software Hadoop,which makes it distributed and parallel.Through experiments and comparison with existing work,it is proved that our method can recognize the social network flow efficiently,and the accuracy is significantly improved.