Incremental Clustering Algorithm of Data Stream Based on Artificial Immune Network

Artificial immune network (aiNet) is an algorithm of artificial immune systems, which is based on the principles of immune system to represent the results of data summarization and clustering, a new incremental clustering algorithm of data stream based on aiNet is proposed in the paper, instead of attempting to directly restructure the clusters, the new approach uses data summarization technique to reflect the dynamic changes of data streams, it has the ability of allowing dynamic tracking of ever-increasing large scale information without having to perform complete re-clustering. The results of experimental evaluation show that the new approach provides significantly faster data summarization than complete re-clustering does. Based on losing little quality of clustering, it has significant superiority on running time. So it can meet the needs of solving the on-line large scale dynamic data clustering in different applications of data streams