The Models of Similarity and Association

1. Auflage 1996
ISBN: 978-0-8039-5707-7
Verlag: Sage Publications

Buch, Englisch, Band 112, 72 Seiten, Format (B × H): 140 mm x 216 mm, Gewicht: 104 g

Reihe: Quantitative Applications in the Social Sciences

Clustering and tree models are widely used in the social and biological sciences to analyze similarity relations. Tree Models of Similarity and Association describes how matrices of similarities or associations among entities can be modeled using trees, and to explain some of the issues that arise in performing such analyses and correctly interpreting the results. James E. Corter clearly distinguishes ultrametric trees (fit by the techniques widely known as "hierarchical clustering") from additive trees and discusses how specific aspects of each type of tree can be interpreted through the use of applications as examples. He concludes with a discussion of when tree models might be preferable to spatial geometric models, such as those fit by multidimensional scaling (MDS) or principal components analysis (PCA).

Autoren/Hrsg.

Corter, James E.

Fachgebiete

• Sozialwissenschaften Soziologie | Soziale Arbeit Soziologie Allgemein Demographie, Demoskopie
• Mathematik | Informatik Mathematik Numerik und Wissenschaftliches Rechnen Angewandte Mathematik, Mathematische Modelle
• Sozialwissenschaften Soziologie | Soziale Arbeit Soziologie Allgemein Gesellschaftstheorie
• Mathematik | Informatik Mathematik Numerik und Wissenschaftliches Rechnen Computeranwendungen in der Mathematik

Inhaltsverzeichnis

Introduction Two Types of Tree Models Algorithms for Fitting Trees to Data Practical Issues and Applications Some Extensions of Tree Models Discussions and Conclusions Appendix A: Mathematical Programming Appendix B: Availability of Software for Fitting Trees Appendix C: Estimating Fit of a Tree Using Multiple Regression

Über Autor(innen)

Corter, James E.
Scholarly Interests

Computational models of human learning and categorization.  Judgment and decision-making. Clustering and scaling methods for multivariate data.  Statistics expertise and probability problem-solving.  Evaluation of educational technology innovations.