A preliminary inquiry into the aggregation of compositional data

DOI

10.2991/eusflat-19.2019.27How to use a DOI?

Keywords

Aggregation Compositional data Penalty functions Weighted centroid

Abstract

The aggregation of compositional data vectors arises naturally in many fields of application. In particular, the weighted centroid is inherently linked to the process of mixing two or more compounds. However, as natural and simple as this process may be, the weighted centroid does not fit the standard definition of an aggregation function since there is no order relation, nor associated monotonicity, in the setting of compositional data. In this conference paper, we prove that the weighted centroid fits within the framework of penalty-based aggregation (which does not build upon the property of monotonicity) and discuss how all convex hull internal functions for the aggregation of multidimensional data could potentially be used for the aggregation of compositional data. Interestingly, the weighted centroid is indeed monotone when seen as a convex hull internal function for the aggregation of multidimensional data.

Copyright

© 2019, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Raúl Pérez-Fernández
AU  - Bernard De Baets
PY  - 2019/08
DA  - 2019/08
TI  - A preliminary inquiry into the aggregation of compositional data
BT  - Proceedings of the 11th Conference of the European Society for Fuzzy Logic and Technology (EUSFLAT 2019)
PB  - Atlantis Press
SP  - 178
EP  - 183
SN  - 2589-6644
UR  - https://doi.org/10.2991/eusflat-19.2019.27
DO  - 10.2991/eusflat-19.2019.27
ID  - Pérez-Fernández2019/08
ER  -