The stability of two common matching functions in classification with respect to a proposed measure

A measure for the quantification of the changes in classification under small changes in data is proposed. A comparison of two common matching functions, namely the single matching function and the cosine function, is made with respect to the measure. The sensitivities of clusters defined as connect...

Full description

Saved in:
Bibliographic Details
Published inJournal of the American Society for Information Science Vol. 27; no. 4; pp. 248 - 255
Main Author Yu, C. T.
Format Journal Article
LanguageEnglish
Published Washington, D.C Wiley Subscription Services, Inc., A Wiley Company 01.07.1976
American Documentation Institute
Wiley Periodicals Inc
Subjects
Algorithms
Automatic classification
Automatic subject indexing
Classification
Clustering
Comparative studies
Correlation
Cosine correlation coefficient
Graph theory
Information storage and retrieval
Information work
Matching
Matching functions
R&D
Research & development
Similarity measures
Simple matching
Stability
Statistical techniques
Subject indexing
Taxonomy
Technical services
Trigonometric functions
Online AccessGet full text
ISSN0002-8231
1097-4571
DOI10.1002/asi.4630270410

Cover

More Information
Summary:A measure for the quantification of the changes in classification under small changes in data is proposed. A comparison of two common matching functions, namely the single matching function and the cosine function, is made with respect to the measure. The sensitivities of clusters defined as connected components and as maximal complete subgraphs are also compared.
Bibliography:istex:E3E7CC8E0FF789F63FD852087651CAB4ACFBFFB9
ArticleID:ASI4630270410
ark:/67375/WNG-KKWNZZHQ-J
Canadian National Research Council
ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 14
ObjectType-Statistics/Data Report-1
ObjectType-Article-1
ObjectType-Feature-2
content type line 23
ISSN:0002-8231
1097-4571
DOI:10.1002/asi.4630270410