A new clustering algorithm for uncertain data stream based on the existence probability of tuple

作者：Ju Chunhua^*; Zhang Pengkun

来源：International Review on Computers and Software, 2012, 7(3): 1038-1044.

• Clustering
• Existence strength
• Uncertain data streams

In this paper, we proposed a new clustering algorithm called UDSC for uncertain data stream. According to the existence probability of cluster, clusters are divided into three levels: strong, transitional and weak. An effective strategy for choosing candidate cluster is developed based on four factors: distance, cluster compactness, existence probability of cluster and existence strength of cluster, which used to find optimal accepted cluster for each continuously arriving data. Then we proposed several rules to describe drifts and shifts in clusters, which used to help users known the changes clearly in clusters and help users make new decisions accurately. Experiment results shows that UDSC clustering algorithm outperforms the existing methods in efficiency and effectiveness.

• 出版日期2012