Fundamentals of Statistics contains material of various lectures and courses of H. Lohninger on statistics, data analysis and chemometrics......click here for more. 
Home Multivariate Data Modeling Classification and Discrimination Cluster Analysis kMeans Clustering  
See also: KNN classification  
kMeans ClusteringKmeans clustering is one of the simplest exchange methods. It may be used for larger datasets as it is rather quick if only a few cluster centers are involved. Kmeans clustering requires the number of clusters k to be specified by the user.From a mathematical viewpoint kmeans clustering is equivalent to an minimisation of the target function where is the distance between the datapoint i and the cluster center j.
Algorithm
Disadvantages of the kmeans algorithm Apart from the simple and quick implementation the kmeans procedure exhibits a few disadvantages which must not be overlooked:


Home Multivariate Data Modeling Classification and Discrimination Cluster Analysis kMeans Clustering 