A Fast Clustering Algorithm for Massive Short Message

Ya Huang, Wenzhi Zhang, Haiyang Zhang, Saihong Xu

Abstract: With the rapid development of mobile communication technology, the short message is playing a more and more important role in the daily life. Most of existing clustering algorithms are hard to be applied in dealing with massive short message due to the huge scale of data and similarity. This paper presents an efficient clustering algorithm by taking a special method to build feature string and a reasonable selection of cluster number. Experiments show that the clustering system based on this algorithm can depose millions of short message per hour with high precision and recall.