Behaviormetrika
Online ISSN : 1349-6964
Print ISSN : 0385-7417
ISSN-L : 0385-7417
Volume 22 , Issue 2
Showing 1-5 articles out of 5 articles from the selected issue
  • Mika Sato, Yoshiharu Sato
    1995 Volume 22 Issue 2 Pages 115-128
    Published: 1995
    Released: May 22, 2006
    JOURNALS RESTRICTED ACCESS
    The purpose of this paper is to propose a generalized clustering model. The structure of an observed similarity is usually unknown and complicated, so various fuzzy clustering models are required to identily the latent structure of the similarity data. Therefore, we define a general class of fuzzy clustering models, so as to represent many different structures of a similarity data. We have been discussed the additive fuzzy clustering model (Sato and Sato, 1994a, 1994b). The merits of the fuzzy clustering models are 1) the amount of computations for the identification of the models are much fewer than in a hard clustering model and 2) we obtain a suitable fitness by a using fewer number of clusters (Sato and Sato, 1994c). In the generalized clustering model, aggregation operators are used to define the degree of simultaneous belongingness of a pair of objects to a cluster. We will discuss some required conditions for the aggregation operators. T-norms are concrete examples for satisfying these conditions. Moreover, the validity of this model is shown by investigating a characteristic of the model and its numerical applications.
    Download PDF (506K)
  • Noriyuki Matsuda, Miki Namatame
    1995 Volume 22 Issue 2 Pages 129-143
    Published: 1995
    Released: May 22, 2006
    JOURNALS RESTRICTED ACCESS
    A new scaling technique was developed in the present work to investigate the image integration of consumers, taking advantage of the recent visual and object-oriented programming technology. Employed visual stimuli were the logos (and the accompanying symbolic marks, if any) of the four leading cosmetic makers, the logos of their brands and the photos of their product packages of skincare items targeted at young females. 19 females subjects evaluated the stimuli with respect to four, five and eight items for the corporate (Clmg), the brand (Blmg) and the package (Plmg) images. The image inheritance was analyzed on the internally anchored items. In addition, external anchores were employed to facilitate the Kansei processing in part of the BImg evaluations. The major findings included the contrasting patterns exhibited by two makers both in the maker/brand discrepancies and in the image inheritances. Among the important issues to be considered in the future work are the mediation effect of the matching task and the class-structure of evaluative items like those employed in AHP.
    Download PDF (725K)
  • Hisao Miyano
    1995 Volume 22 Issue 2 Pages 145-153
    Published: 1995
    Released: May 22, 2006
    JOURNALS RESTRICTED ACCESS
    In this paper, we propose a new hierarchical clustering method, which is useful to find appropriate clusters of attributes from given dichotomous or frequency data. Important features of our method are 1) the similarity between two attributes is defined as a probability of their pattern vectors being observed under the hypothesis of independence, 2) for each generated cluster, one pattern vector is defined in a natural manner, and 3) it can be used freely without distinguishing the frequency data from the dichotomous one. A typical frequency data is analyzed to illustrate how our method works effectively. The discussion on similarities among objects is also included to propose a new similarity measure based on our clustering method.
    Download PDF (415K)
  • Ryozo Yoshino, Diana Khor
    1995 Volume 22 Issue 2 Pages 155-184
    Published: 1995
    Released: May 22, 2006
    JOURNALS RESTRICTED ACCESS
    This paper discusses different methods of measuring and comparing national character cross-nationally, considered respectively in terms of a multifaceted phenomenon and a specific aspect such as the quality of life. Using data collected in seven nations, we first applied the GHT model with VA representations to both a global and a specific pool of quality of life items, making use of the concept of “superculture” as a basis of comparison among countries. We then undertook principal components analysis for another global pool of items and the quality of life items, comparing the structure of different attitudes and the substantive level of quality of life among countries. The substantive results of the different methods of analyses concurred in some respects and differed in others. Importantly, the paper underscored the complementarity of different methods as equally fruitful ways to understanding national character in a cross-national context.
    Download PDF (1583K)
  • Kenichi Kikuchi, Shin-ichi Mayekawa
    1995 Volume 22 Issue 2 Pages 185-204
    Published: 1995
    Released: May 22, 2006
    JOURNALS RESTRICTED ACCESS
    When we are to select students using the result of an entrance examination, it is common to select N0 students out of N applicants on the basis of a composite score of m subtests, where N0, the number of students to be selected, is determined prior to the administration of the entrance examination. The swap-rate is an intuitively easy-to-understand measure of the contribution of each subtest to the selection. The swap-rate of the j-th test is defined as the proportion, to N0, of the number of students who failed to pass the examination if the j-th test is not used for the selection. In this paper we first discuss the sampling distribution of the swap-rate under the normality assumption and then show the result of a bootstrapping using real selection data.
    Download PDF (953K)
feedback
Top