Program clustering k-means. The archive contains a project consisting of 6 classes.
ClusteringKMeans - a class that implements the k-means
DataLoader - Data Loader
DataProcessing - class with statistical functions
DataShell - a class that represents a shell to work with clustering methods
Vector - a class of special functions for manipulating vectors and matrices
Program - the main class, the entry point
Interface for use with the shell consists of methods:
public void Load (string path, int countComponents) - download data for clustering file (path - the path, countComponent - the number of components of the sample)
public void Predprocessing () - pre-processing of data (normalization of samples of each component in order to reduce the influence of large absolute value components of the sample)
public void Clustering (int countClusters) - the actual start clustering (countClusters - the number of clusters)
public void ShowSamplesInClusters () - displays the results of clustering
Caution To work correctly, you must prescribe the correct path to the text file with the data! The data in the file are represented as strings with the entered value of each component separated by a space. For example:
This console program!