Monitoring Changes in Clustering Solutions: A Review of Models and Applications

Table 1

Clustering algorithms and its types.

Class	Input parameters	Example	Performance evaluation
Partitioning	Number of clusters (k)	k-Means, HCL, CLARA	Dunn, Davies-Bouldin, Rand
	Distance function	k-Mediod, neural gas	Calinski-Harabasz, Jaccard
Hierarchical	Distance function	CURE, BIRCH, ROCK	Cophenetic correlation
	Linkage function	CHAMELEON	Dendrogram-based measures
Density-based	Epsilon	DBSCAN, LDBSCAN, OPTICS	Density reachability F, CSI
	Minimum points	DENCLUE, STDBSCAN	Adjusted mutual information
Grid-based	No. of grid cells	STING, CLIQUE	Grid quality index, grid entropy
	The wavelet	WaveCluster	Grid dispersion, grid purity
Model-based	No. of components	GMM	Akaike information criterion
	Model parameters		Bayesian information criterion