[en] In this paper methods to cluster analyze two-mode data are discussed
which assume that both objects and attributes contribute to the uncovering of
meaningful patterns of clusters. Two-mode methods are reviewed and criteria are
proposed which aim at a comparison and evaluation of the reviewed methods. The
selected criteria show that most two-mode approaches su®er from drawbacks con-
cerning interpretation of the data, convergence of algorithms, uniqueness of solu-
tions or applicability to larger data sets. They imply some suggestions for future
directions in the development of two{ and three{mode cluster analysis.
Disciplines :
Social & behavioral sciences, psychology: Multidisciplinary, general & others
Identifiers :
UNILU:UL-CHAPTER-2011-136
Author, co-author :
Krolak-Schwerdt, Sabine ; University of Luxembourg > Faculty of Language and Literature, Humanities, Arts and Education (FLSHASE) > Languages, Culture, Media and Identities (LCMI)
Language :
English
Title :
Two-mode methods of cluster analysis: Compare and contrast
Publication date :
2003
Main work title :
Studies in classification, data analysis, and knowledge organization
Editor :
Schader, M.
Gaul, W.
Vichi, M.
Publisher :
Springer, Berlin, Unknown/unspecified
Collection name :
Vol. 24: Between data science and applied data analysis