Combinatorial K-means clustering

Designed by Dr. Sergio Madurga

Materials Science and Physical Chemistry Department & Research Institute of Theoretical and Computational Chemistry (IQTCUB), University of Barcelona, C/Martí i Franquès, 08028 Barcelona, Spain; [email protected]

A new original k-means procedure is developed to identify the descriptors that produce the clustering with the optimal separation among groups. It is used a combinatorial procedure to make the selection of the descriptors that minimize the objective parameted colled global variance. A main characteristic of this procedure is that it is not required that all the variables to be known for all the objects. It has been tested that this technique could be applied to medical clinical data of patients with Diabetes Mellitus Type 2 (DMT2) with underlying diseases.

Citation: Nedyalkova, M.; Madurga, S.; Simeonov, V. Combinatorial K-means clustering as a machine learning tool applied to diabetes mellitus type 2. Int. J. Environ. Res. Public Health 2021 Feb 17;18(4):1919. doi: 10.3390/ijerph18041919.