Jenks Natural Breaks

labels/tags

Tagged according to tag list
clustering

Principle

The Jenks optimization method, also known as the Jenks natural breaks classification method, is a monovariate numerical data clustering technique which determine the best arrangement of values in cluster. It consists to minimize each cluster’s average deviation compare to its mean. Principle is to maximize each cluster’s deviation from the means with other classes. It can be seen as a precessus which seeks to minimize the variance within clusters and maximize the variance between them. It can be considered as a particular case of $K$-Means for monovariate numerical data.

Such as $K$-Means it takes the $K$ argument to ask how many clusters are expected. Contrary to $K$-Means it doesn’t require other arguments.

scalability

Complexity is in $O(…)$.

Input

A collection of numerical monovariate data.

parameters

$K$ the number of expected clusters.

Ouput format

The model consist of the ascendant ordered numerical break values.