Bibliographic coupling and hierarchical clustering for the validation and improvement of subject-classification schemes

Bart Thijs, Lin Zhang, W. Glänzel

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

An attempt is made to cluster journals from the complete Web of Science database by using bibliographic coupling similarities. Since the sparseness of the underlying similarity matrix proved inappropriate for this exercise, second-order similarities have been used. Only 0.12 % out of 8282 journals had to be removed from the classification as being singletons. The quality at three hierarchical levels with 6, 14 and 24 clusters substantiated the applicability of this method. Cluster labelling was made on the basis of the about 70 subfields of the Leuven–Budapest subject-classification scheme that also allowed the comparison with the existing two-level journal classification system developed in Leuven. The further comparison with the 22 field classification system of the Essential Science Indicators does, however, reveal larger deviations.

Original languageEnglish
Pages (from-to)1453-1467
Number of pages15
JournalScientometrics
Volume105
Issue number3
DOIs
Publication statusPublished - Dec 1 2015

Fingerprint

science
Labeling

Keywords

  • Bibliographic coupling
  • Journal clustering
  • Second order similarities
  • Subject classification

ASJC Scopus subject areas

  • Computer Science Applications
  • Social Sciences(all)
  • Library and Information Sciences
  • Law

Cite this

Bibliographic coupling and hierarchical clustering for the validation and improvement of subject-classification schemes. / Thijs, Bart; Zhang, Lin; Glänzel, W.

In: Scientometrics, Vol. 105, No. 3, 01.12.2015, p. 1453-1467.

Research output: Contribution to journalArticle

@article{86f089fe8b94409e87dbf1f282b62fd7,
title = "Bibliographic coupling and hierarchical clustering for the validation and improvement of subject-classification schemes",
abstract = "An attempt is made to cluster journals from the complete Web of Science database by using bibliographic coupling similarities. Since the sparseness of the underlying similarity matrix proved inappropriate for this exercise, second-order similarities have been used. Only 0.12 {\%} out of 8282 journals had to be removed from the classification as being singletons. The quality at three hierarchical levels with 6, 14 and 24 clusters substantiated the applicability of this method. Cluster labelling was made on the basis of the about 70 subfields of the Leuven–Budapest subject-classification scheme that also allowed the comparison with the existing two-level journal classification system developed in Leuven. The further comparison with the 22 field classification system of the Essential Science Indicators does, however, reveal larger deviations.",
keywords = "Bibliographic coupling, Journal clustering, Second order similarities, Subject classification",
author = "Bart Thijs and Lin Zhang and W. Gl{\"a}nzel",
year = "2015",
month = "12",
day = "1",
doi = "10.1007/s11192-015-1641-3",
language = "English",
volume = "105",
pages = "1453--1467",
journal = "Scientometrics",
issn = "0138-9130",
publisher = "Springer Netherlands",
number = "3",

}

TY - JOUR

T1 - Bibliographic coupling and hierarchical clustering for the validation and improvement of subject-classification schemes

AU - Thijs, Bart

AU - Zhang, Lin

AU - Glänzel, W.

PY - 2015/12/1

Y1 - 2015/12/1

N2 - An attempt is made to cluster journals from the complete Web of Science database by using bibliographic coupling similarities. Since the sparseness of the underlying similarity matrix proved inappropriate for this exercise, second-order similarities have been used. Only 0.12 % out of 8282 journals had to be removed from the classification as being singletons. The quality at three hierarchical levels with 6, 14 and 24 clusters substantiated the applicability of this method. Cluster labelling was made on the basis of the about 70 subfields of the Leuven–Budapest subject-classification scheme that also allowed the comparison with the existing two-level journal classification system developed in Leuven. The further comparison with the 22 field classification system of the Essential Science Indicators does, however, reveal larger deviations.

AB - An attempt is made to cluster journals from the complete Web of Science database by using bibliographic coupling similarities. Since the sparseness of the underlying similarity matrix proved inappropriate for this exercise, second-order similarities have been used. Only 0.12 % out of 8282 journals had to be removed from the classification as being singletons. The quality at three hierarchical levels with 6, 14 and 24 clusters substantiated the applicability of this method. Cluster labelling was made on the basis of the about 70 subfields of the Leuven–Budapest subject-classification scheme that also allowed the comparison with the existing two-level journal classification system developed in Leuven. The further comparison with the 22 field classification system of the Essential Science Indicators does, however, reveal larger deviations.

KW - Bibliographic coupling

KW - Journal clustering

KW - Second order similarities

KW - Subject classification

UR - http://www.scopus.com/inward/record.url?scp=84947126842&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84947126842&partnerID=8YFLogxK

U2 - 10.1007/s11192-015-1641-3

DO - 10.1007/s11192-015-1641-3

M3 - Article

VL - 105

SP - 1453

EP - 1467

JO - Scientometrics

JF - Scientometrics

SN - 0138-9130

IS - 3

ER -