Symbolic data analysis is based on special descriptions of data known as symbolic objects (SOs). Such descriptions preserve more detailed information about units and their clusters than the usual representations with mean values. A special type of SO is a representation with frequency or probability distributions (modal values). This representation enables us to simultaneously consider variables of all measurement types during the clustering process. In this paper, we present the theoretical basis for compatible leaders and agglomerative clustering methods with alternative dissimilarities for modal-valued SOs. The leaders method efficiently solves clustering problems with large numbers of units, while the agglomerative method can be applied either alone to a small data set, or to leaders, obtained from the compatible leaders clustering method. We focus on (a) the inclusion of weights that enables clustering representatives to retain the same structure as if clustering only first order units and (b) the selection of relative dissimilarities that produce more interpretable, i.e., meaningful optimal clustering representatives. The usefulness of the proposed methods with adaptations was assessed and substantiated by carefully constructed simulation settings and demonstrated on three different real-world data sets gaining in interpretability from the use of weights (population pyramids and ESS data) or relative dissimilarity (US patents data).
This work was done during authors’ visit to Kavli IPMU and was supported by the World Premier International Research Center Initiative (WPI Initiative), MEXT, Japan. The reported study was partially supported by RFBR, Research Projects 13-01-00234, 14-01-00416 and 15-51-50045. The article was prepared within the framework of a subsidy granted to the HSE by the Government of the Russian Federation for the implementation of the Global Competitiveness Program.
For a left coherent ring A with every left ideal having a countable set of generators, we show that the coderived category of left A-modules is compactly generated by the bounded derived category of finitely presented left A-modules (reproducing a particular case of a recent result of Št’ovíček with our methods). Furthermore, we present the definition of a dualizing complex of fp-injective modules over a pair of noncommutative coherent rings A and B, and construct an equivalence between the coderived category of A-modules and the contraderived category of B-modules. Finally, we define the notion of a relative dualizing complex of bimodules for a pair of noncommutative ring homomorphisms (Formula presented.) and (Formula presented.), and obtain an equivalence between the R / A-semicoderived category of R-modules and the S / B-semicontraderived category of S-modules. For a homomorphism of commutative rings (Formula presented.), we also construct a tensor structure on the R / A-semicoderived category of R-modules. A vision of semi-infinite algebraic geometry is discussed in the introduction.
Let Ψ be the projectivization (i.e., the set of one-dimensional vector subspaces) of a vector space of dimension ≥ 3 over a field. Let H be a closed (in the pointwise convergence topology) subgroup of the permutation group GΨ of the set Ψ. Suppose that H contains the projective group and an arbitrary self-bijection of Ψ transforming a triple of collinear points to a non-collinear triple. It is well-known from  that if Ψ is finite then H contains the alternating subgroup AΨ of GΨ.
We show in Theorem 3.1 below that H = GΨ, if Ψ is infinite.