Title: On the use of valued action profiles for relational multi-criteria clustering

Authors: Stefan Eppe; Julien Roland; Yves De Smet

Addresses: Computer and Decision Engineering Department, Ecole polytechnique de Bruxelles, Université libre de Bruxelles, Boulevard du Triomphe, CP210/01, 1050 Brussels, Belgium ' Computer and Decision Engineering Department, Ecole polytechnique de Bruxelles, Université libre de Bruxelles, Boulevard du Triomphe, CP210/01, 1050 Brussels, Belgium ' Computer and Decision Engineering Department, Ecole polytechnique de Bruxelles, Université libre de Bruxelles, Boulevard du Triomphe, CP210/01, 1050 Brussels, Belgium

Abstract: Clustering techniques aim at eliciting hidden structures of a dataset by partitioning it into groups of similar elements. We will focus on the special case of relational clustering, where the similarity is based on relations that exist between elements rather than on their respective intrinsic features. As will be shown, particular attention has to be paid when applying a relational approach to the context of multi-criteria decision making. This paper introduces the concept of valued action profiles, a formalism for handling elements that are defined by pairwise valued outranking relations. For our experimental study, we then integrate these profiles into an adapted k-means algorithm that returns a relational partition. Results on both artificial and real datasets show that the use of the proposed method leads to meaningful relational partitions.

Keywords: multiple criteria analysis; relational clustering; action profiles; outranking relations; fuzzy relations; MCDM; multicriteria decision making; fuzzy logic; pairwise comparison; valued outranking relations; adapted k-means algorithm; artificial datasets; real datasets; relational partitions.

DOI: 10.1504/IJMCDM.2014.064796

International Journal of Multicriteria Decision Making, 2014 Vol.4 No.3, pp.201 - 233

Received: 02 May 2012
Accepted: 11 Apr 2013

Published online: 15 Sep 2014 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article