The grouping of data by clustering generally depends on the clustering method used and its specific parameter settings. Therefore, the comparison of results obtained from different clusterings is generally recommended (ensemble clustering). The present study presents a simple and an optimized method for visualizing such results by drawing a two-dimensional color map that associates data with cluster memberships. The methodology is applicable to any unsupervised and supervised classification results.
CITATION STYLE
Advances in Data Analysis. (2010). Advances in Data Analysis. Birkhäuser Boston. https://doi.org/10.1007/978-0-8176-4799-5
Mendeley helps you to discover research relevant for your work.