Title: R scripts for automated K-means clustering
Authors: Sung-Soo Kim - Korea National Open University (Korea, South) [presenting]
Abstract: The aim is to provide R scripts for an automated K-means clustering process which decide the number of clusters and the initial cluster center, which is combined with variable selection process in a forward manner. We also provide potential outliers whenever a new variable is added in a K-means clustering process. This process can be used to see the inter-relationship between variables and outliers, and interactive data visualization is combined to clarify the clustering structure and help to identify outliers.