Use the KMeansClusterAnalysis class to perform a K-means cluster analysis of numerical data.

K-means clustering partitions a set of observations into a predefined number of clusters. A number of initial cluster centers is chosen. The observations are assigned to the closest cluster. Each center is then recalculated as the mean of its members. This changes the distances between cluster centers and observations, so the observations are once again reassigned. This process is repeated until no more observations change cluster. Note that the final partition depends on the initial location of the centers. Different applications may return different results for the same dataset.

The number of clusters must be passed in a constructor. It can be accessed through the NumberOfClusters property. The Standardize property determines whether the variables are transformed to all have the same mean and standard deviation. The default is trueTruetruetrue (True in Visual Basic).

Once the clustering is computed by calling the Compute(ParallelOptions) method, the GetClusters()()()() method gives access to a KMeansClusterCollection that gives detailed information about the results. In addition to the usual collection properties and methods, it has two interesting methods: GetMemberships()()()() returns a CategoricalVariable that for each observation indicates the cluster it belongs to. GetDistancesToCenters()()()() returns a NumericalVariable that contains the distance of each observation to the center of its cluster.

Numerical Libraries