Hybrid clustering for validation and improvement of subject-classification schemes

Janssens, Frizo; Zhang, Lin; Moor, Bart De; Glänzel, Wolfgang

doi:10.1016/j.ipm.2009.06.003

Cited by 83 publications

(42 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A more appropriate method is based on Fisher's inverse v 2 -method (cf. Janssens et al 2009). The linear combination of angles forms an interesting alternative to Fisher's inverse v 2 -method for handling similarities from different metrics and offers the opportunity of smoothly adjusting the weight of either component.…”

Section: Methodsmentioning

confidence: 99%

Using ‘core documents’ for the representation of clusters and topics

Glänzel

Thijs

2011

Scientometrics

Self Cite

View full text Add to dashboard Cite

The notion of 'core documents', first introduced in the context of co-citation analysis and later re-introduced for bibliographic coupling, refers to the representation of the core of a publication set according to given criteria. In the present study, the notion of core documents is extended to the combination of citation-based and textual links. It is shown that core documents defined this way can be used to represent and describe document clusters and topics at different levels of aggregation. Methodology is illustrated using the example of two ISI Subject Categories selected from applied and social sciences.

show abstract

Section: Methodsmentioning

confidence: 99%

Using ‘core documents’ for the representation of clusters and topics

Glänzel

Thijs

2011

Scientometrics

Self Cite

View full text Add to dashboard Cite

show abstract

“…The PageRank of a journal can be understood here as the probability that a random reader will be reading that journal, when he/she randomly and with equal probability looks up cited references to other journals (different from the current one), but once in a while randomly picks another journal from the library (cluster). In general, the journals ranking highest represent their cluster in an adequate manner (Janssens et al 2009). …”

Section: Methodsmentioning

confidence: 99%

“…''Good migration'' is observed if the goodness of the unit's classification improves, and based on its title and scope, clearly it should indeed be assigned to the cluster to which it has moved. Otherwise we call it ''bad migration'' (Janssens et al 2009). According to the scope of the journals, these migrations are not convincing.…”

Section: Detecting ''Migration'' Of Journals and Improving The ''Intementioning

confidence: 99%

Journal cross-citation analysis for validation and improvement of journal-based subject classification in bibliometric research

et al. 2010

Self Cite

View full text Add to dashboard Cite

The objective of this study is to use a clustering algorithm based on journal cross-citation to validate and to improve the journal-based subject classification schemes. The cognitive structure based on the clustering is visualized by the journal cross-citation network and three kinds of representative journals in each cluster among the communication network have been detected and analyzed. As an existing reference system the 15-field subject classification by Glänzel and Schubert (Scientometrics 56:55-73, 2003) has been compared with the clustering structure.

show abstract

“…Beyond classification of journals into fields and subfields, the auxiliary aim of the experiment was to explore the relationship of this novel classification to existing ones, of which the most straightforward and first example is the system of subject categories used by the Essential Science Indicators (ESI) database of ThomsonReuters (formerly, ISI). In this respect, our approach was related to a recent study of Janssens et al (2009), where the results of a hybrid clustering technique, based on both textual characteristics and cross-citation patterns of journals, were contrasted with the ESI classification scheme. ESI classifies the journals covered by the Thomson-Reuters databases into 22 categories (see Table 1).…”

Section: Methodsmentioning

confidence: 99%