the dataset
number of clusters
distance between ancient modes and update modes under which we consider than the algorithm have converged, if and only if all every mode have converged
number maximal of iteration
the dissimilarity measure used
The modes of each cluster.
Dimension
Return the nearest mode for a specific point.
Return the nearest mode for a specific point.
K-Modes clustering. K-Modes is the binary equivalent for K-Means. The mean update for centroids is replace by the mode one which is a majority vote among element of each cluster.
Complexity is O(k.t.n) with Hamming distance, quadratic with other metrics.
References