2007
DOI: 10.1002/asi.20732
|View full text |Cite
|
Sign up to set email alerts
|

On the normalization and visualization of author co‐citation data: Salton's Cosine versus the Jaccard index

Abstract: The debate about which similarity measure one should use for the normalization in the case of Author Co-citation Analysis (ACA) is further complicated when one distinguishes between the symmetrical co-citation-or, more generally, co-occurrence-matrix and the underlying asymmetrical citation-occurrence-matrix. In the Web environment, the approach of retrieving original citation data is often not feasible. In that case, one should use the Jaccard index, but preferentially after adding the number of total citatio… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

3
130
1
10

Year Published

2008
2008
2017
2017

Publication Types

Select...
8
1

Relationship

1
8

Authors

Journals

citations
Cited by 189 publications
(147 citation statements)
references
References 21 publications
3
130
1
10
Order By: Relevance
“…As in many other studies examining scholarly collaboration (Luukkonen et al 1993;Leydesdorff 2008), the Jaccard index was used for normalisation purposes. Where it seemed appropriate and helpful, we used visualisation to present the outcome of our analyses drawing on the Pajek (2008) software package, applying the Kamada-Kawai (1989) algorithm to our small datasets (Nooy et al 2005).…”
Section: Methodsology and Data Collectionmentioning
confidence: 99%
“…As in many other studies examining scholarly collaboration (Luukkonen et al 1993;Leydesdorff 2008), the Jaccard index was used for normalisation purposes. Where it seemed appropriate and helpful, we used visualisation to present the outcome of our analyses drawing on the Pajek (2008) software package, applying the Kamada-Kawai (1989) algorithm to our small datasets (Nooy et al 2005).…”
Section: Methodsology and Data Collectionmentioning
confidence: 99%
“…Las posibilidades de ésta técnica continúan ampliándose con los trabajos de Garfi eld (1994), sobre mapeo longitudinal para detectar los avances de la investigación científi ca en un orden cronológico, y con los diferentes trabajos de Noyons sobre mapas bibliométrico para medir la similitud de publicaciones por medio palabras claves (Noyons y Van Raan, 1998;Noyons y otros, 1999;Leydesdorff y Wouters, 1999). Durante la primera década del nuevo siglo, se han dispuesto nuevas herramientas de trabajo para el ACA con la implementación de diferentes técnicas de normalización para matrices de autores y la utilización de variadas técnicas de visualización de datos como ocurre con el caso de las redes Pathfi nder (Chen, 1998;White, 2003a;Chen y Kuljis, 2003;Leydesdorff, 2008;Egghe y Leydesdorff, 2009 …”
Section: Análisis De Cocitación De Autor (Aca)unclassified
“…Para estos autores, el coefi ciente de Pearson ha representado una excelente medida de similaridad que permite identifi car la cercanía de perfi les de cocitación y eliminar las diferencias entre escalas de aquellos autores que son muy citados (Griffi th, 1980;White y Griffi th, 1981;McCain, 1990;White y McCain, 1998;White, 2003b). A pesar de que ha sido ampliamente aceptado en el ACA como medida de similaridad, el r de Pearson ha sido cuestionado como medida de distancia en la visualización de la matriz de datos debido a su sensibilidad a la inclusión de ceros en la matriz (Ahlgren y otros, 2003;Leydesdorff y Vaughan, 2006;Leydesdorff, 2008;Egghe y Leydesdorff, 2009). Para solucionar esta defi ciencia, Egghe y Leydesdorff (2009) han propuesto relacionar el coefi ciente de Pearson con el coseno de Salton por medio de un umbral de transformación, mientras que White (2003a), ha propuesto utilizar el algoritmo PFNET del mínimo peso entre dos puntos, por medio de la defi nición de los parámetros r y q de Chen (1998).…”
Section: Matrices De Cocitaciónunclassified
“…Egghe and Leydesdorff [15] revealed the relation between Pearson's correlation coefficient and Salton's cosine measure based on the different possible values of the division of the -norm and the -norm of a vector; where they analysed author co-citations among 24 informetricians and constructed two matrices: the asymmetric occurrence matrix and the symmetric co-citation matrix; and got a threshold value for the cosine of which none of the corresponding Pearson correlations would be negative. Leydesdorff [21] argued that in the web environment, the approach of retrieving original citation data is often not feasible and in that case, one should use the Jaccard index. Unlike Salton's cosine and the Pearson correlation, the Jaccard index abstracts from the shape of the distributions and focuses only on the intersection and the sum of the two sets.…”
Section: Introductionmentioning
confidence: 99%