Premium
Multicriteria Classification with Unknown Categories: A Clustering–Sorting Approach and an Application to Conflict Management
Author(s) -
Rocha Clara,
Dias Luis C.,
Dimas Isabel
Publication year - 2012
Publication title -
journal of multi‐criteria decision analysis
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.462
H-Index - 47
eISSN - 1099-1360
pISSN - 1057-9214
DOI - 10.1002/mcda.1476
Subject(s) - sorting , cluster analysis , sort , heuristic , computer science , set (abstract data type) , similarity (geometry) , order (exchange) , decision maker , measure (data warehouse) , data mining , simple (philosophy) , artificial intelligence , machine learning , mathematics , image (mathematics) , operations research , algorithm , information retrieval , epistemology , economics , philosophy , finance , programming language
This work proposes an approach to cluster and sort a set of alternatives considering multi‐criteria categories with a partial order structure. It can be considered a heuristic approach because it does not attempt to derive an optimal partial order among all conceivable clusters of alternatives. Rather than this, it intends to be a simple approach that is transparent to the Decision Maker (DM) whose assistance is sought to help shaping the results. The approach proposed arises from the conjugation of traditional Clustering analysis and Multi‐criteria sorting tools. At the outset, the number of categories and their characteristics is unknown. First, we need to detect only the clusters themselves on the basis of a similarity measure independent of the preferences of the DM. Next, we detect potential partial order relations that might exist between them, according to the subjective preferences of the DM. Such preferences are elicited only after the DM has examined the clusters detected and deemed that these categories made sense. The new approach performs very well in a real‐world problem of management of intragroup conflicts and conflict handling strategies. Copyright © 2012 John Wiley & Sons, Ltd.