Data Stream Clustering and Outlier Detection Algorithm Based on Shared Nearest Neighbor Density

A stream clustering and outlier detection algorithm was proposed based on the shared nearest neighbor density. In this algorithm, the shared nearest neighbor density was defined. The shared nearest neighbor density considered the number of the nearest neighbor and the nearest neighbor as other data objects. So the clustering result was not influenced by the density variation. The cluster density were defined to identify outliers. Experimental results demonstrate the algorithm can discover clusters of different densities and distinguish outliers. The improved algorithm has the performance of clustering effect and a better clustering quality.