An incremental clustering algorithm based on swarm intelligence theory
Zhuo Chen, Qingchun Meng · 2005
Although many clustering algorithms have been proposed so far, they seldom focused on high-dimensional and incremental databases. We present a new type of incremental clustering algorithm, which is based on the swarm intelligence theory. The new type of incremental clustering algorithm implements the clustering process by the actions of clustering agents. The clustering agents, which move in a three-dimensional space, have the abilities of memory, communication, analysis, judgement, coordination and so on. This new type of incremental clustering algorithm is applicable in periodically incremental environment. Experimental results have shown that this algorithm has many merits such as insensitivity to the order of the data, capability of dealing with the exceptional, high-dimensional and complicated data.