Data Clustering Using Divisive Analysis Method (Diana) (by: Chandra Gunawan, Dewi Rachmatin, and Maman Suherman)

It was announced at unisba mathematics seminar in 2014


Cluster Analysis is a data grouping analysis that groups data based on the information specified in the data. The purpose of cluster analysis is for objects in one group to have similarities to each other whereas with different objects the groups have differences. The process of analyzing clusters or grouping data can be done with two methods: hierarchy method and non-hierarchical method. The hierarchy method is divided into two, the agglomerative and divisive methods. In this article, the hierarchy method will be discussed in the form of a divisive method.

At each step in the divisive method there is an addition of groups into the values of the two smallest values until finally all the elements are joined. This divisive method is a clustering process based on the average value equation between objects. If an object has the largest average value equation then it will split and turn into a splinter group. At the end of this article will discuss the results of the implementation of divisive methods on ten data on air pollution levels in the United States.