Constrained classification: The use of a priori information in cluster analysis

Wayne Desarbo, Vijay Mahajan

Research output: Contribution to journalArticle

41 Citations (Scopus)

Abstract

In many classification problems, one often possesses external and/or internal information concerning the objects or units to be analyzed which makes it appropriate to impose constraints on the set of allowable classifications and their characteristics. CONCLUS, or CONstrained CLUStering, is a new methodology devised to perform constrained classification in either an overlapping or nonoverlapping (hierarchical or nonhierarchial) manner. This paper initially reviews the related classification literature. A discussion of the use of constraints in clustering problems is then presented. The CONCLUS model and algorithm are described in detail, as well as their flexibility for use in various applications. Monte Carlo results are presented for two synthetic data sets with appropriate discussion of the resulting implications. An illustration of CONCLUS is presented with respect to a sales territory design problem where the objects classified are various Forbes-500 companies. Finally, the discussion section highlights the main contribution of the paper and offers some areas for future research.

Original languageEnglish (US)
Pages (from-to)187-215
Number of pages29
JournalPsychometrika
Volume49
Issue number2
DOIs
StatePublished - Jun 1 1984

Fingerprint

Cluster analysis
Cluster Analysis
Clustering
Synthetic Data
Classification Problems
Overlapping
Flexibility
Internal
Sales
Unit
Methodology
Industry
Object
Model

All Science Journal Classification (ASJC) codes

  • Psychology(all)
  • Applied Mathematics

Cite this

Desarbo, Wayne ; Mahajan, Vijay. / Constrained classification : The use of a priori information in cluster analysis. In: Psychometrika. 1984 ; Vol. 49, No. 2. pp. 187-215.
@article{4fb97764bcc74028be4289c2b3ca3b3e,
title = "Constrained classification: The use of a priori information in cluster analysis",
abstract = "In many classification problems, one often possesses external and/or internal information concerning the objects or units to be analyzed which makes it appropriate to impose constraints on the set of allowable classifications and their characteristics. CONCLUS, or CONstrained CLUStering, is a new methodology devised to perform constrained classification in either an overlapping or nonoverlapping (hierarchical or nonhierarchial) manner. This paper initially reviews the related classification literature. A discussion of the use of constraints in clustering problems is then presented. The CONCLUS model and algorithm are described in detail, as well as their flexibility for use in various applications. Monte Carlo results are presented for two synthetic data sets with appropriate discussion of the resulting implications. An illustration of CONCLUS is presented with respect to a sales territory design problem where the objects classified are various Forbes-500 companies. Finally, the discussion section highlights the main contribution of the paper and offers some areas for future research.",
author = "Wayne Desarbo and Vijay Mahajan",
year = "1984",
month = "6",
day = "1",
doi = "10.1007/BF02294172",
language = "English (US)",
volume = "49",
pages = "187--215",
journal = "Psychometrika",
issn = "0033-3123",
publisher = "Springer New York",
number = "2",

}

Constrained classification : The use of a priori information in cluster analysis. / Desarbo, Wayne; Mahajan, Vijay.

In: Psychometrika, Vol. 49, No. 2, 01.06.1984, p. 187-215.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Constrained classification

T2 - The use of a priori information in cluster analysis

AU - Desarbo, Wayne

AU - Mahajan, Vijay

PY - 1984/6/1

Y1 - 1984/6/1

N2 - In many classification problems, one often possesses external and/or internal information concerning the objects or units to be analyzed which makes it appropriate to impose constraints on the set of allowable classifications and their characteristics. CONCLUS, or CONstrained CLUStering, is a new methodology devised to perform constrained classification in either an overlapping or nonoverlapping (hierarchical or nonhierarchial) manner. This paper initially reviews the related classification literature. A discussion of the use of constraints in clustering problems is then presented. The CONCLUS model and algorithm are described in detail, as well as their flexibility for use in various applications. Monte Carlo results are presented for two synthetic data sets with appropriate discussion of the resulting implications. An illustration of CONCLUS is presented with respect to a sales territory design problem where the objects classified are various Forbes-500 companies. Finally, the discussion section highlights the main contribution of the paper and offers some areas for future research.

AB - In many classification problems, one often possesses external and/or internal information concerning the objects or units to be analyzed which makes it appropriate to impose constraints on the set of allowable classifications and their characteristics. CONCLUS, or CONstrained CLUStering, is a new methodology devised to perform constrained classification in either an overlapping or nonoverlapping (hierarchical or nonhierarchial) manner. This paper initially reviews the related classification literature. A discussion of the use of constraints in clustering problems is then presented. The CONCLUS model and algorithm are described in detail, as well as their flexibility for use in various applications. Monte Carlo results are presented for two synthetic data sets with appropriate discussion of the resulting implications. An illustration of CONCLUS is presented with respect to a sales territory design problem where the objects classified are various Forbes-500 companies. Finally, the discussion section highlights the main contribution of the paper and offers some areas for future research.

UR - http://www.scopus.com/inward/record.url?scp=0000727780&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0000727780&partnerID=8YFLogxK

U2 - 10.1007/BF02294172

DO - 10.1007/BF02294172

M3 - Article

AN - SCOPUS:0000727780

VL - 49

SP - 187

EP - 215

JO - Psychometrika

JF - Psychometrika

SN - 0033-3123

IS - 2

ER -