Bibliometric Routine

Objetivo

Estabelecer um script pronto para análise de bibliometria e cientometria para consultas realizadas no ISI - Web Of Knowledge http://www.webofknowledge.com/

Este script é baseado no pacote “Bibliometrix” http://www.bibliometrix.org/ para R. Na aba Documents é possível baixar um tutorial para preparar os dados para rodar no pacote.

Sendo assim, o script abaixo considera que o usuário já baixou a consulta em .txt ou .bib. A título de demonstração consultamos no ISI o termo “autonomous vehicles” e selecionamos apenas áreas que se refere aos “veículos autonômos” que estão sendo desenvolvidos para o ambiente urbano.

O documento consta apenas os resultados sendo que o script em formato .RMD ou .R pode ser obtido neste link: https://github.com/fabiorcampos/Bibliometric-Routines

 [1] "\n\nMain Information about data\n\n"                 
 [2] "Articles                              291 \n"        
 [3] "Sources (Journals, Books, etc.)       141 \n"        
 [4] "Keywords Plus (ID)                    474 \n"        
 [5] "Author's Keywords (DE)                365 \n"        
 [6] "Period                                1985 - 2015 \n"
 [7] "Average citations per article         11.73 \n\n"    
 [8] "Authors                               537 \n"        
 [9] "Author Appearances                    647 \n"        
[10] "Authors of single authored articles   105 \n"        
[11] "Authors of multi authored articles    432 \n\n"      
[12] "Articles per Author                   0.542 \n"      
[13] "Authors per Article                   1.85 \n"       
[14] "Co-Authors per Articles               2.22 \n"       
[15] "Collaboration Index                   2.94 \n"       
[16] "\n"                                                  

Authors’ Dominance ranking

The function dominance calculates the authors’ dominance ranking as proposed by Kumar & Kumar, 2008.

Authors’ h-index

The h-index is an author-level metric that attempts to measure both the productivity and citation impact of the publications of a scientist or scholar.

authors=gsub(","," ",names(df$Authors)[1:10])

indices <- Hindex(database, authors, sep = ";",years=50)

knitr::kable(indices$H)

Lotka’s Law coefficient estimation

Lotka’s law describes the frequency of publication by authors in any given field as an inverse square law, where the number of authors publishing a certain number of articles is a fixed ratio to the number of authors publishing a single article.

L <- lotka(df)
# Author Productivity. Empirical Distribution
knitr::kable(L$AuthorProd)

Bibliometric network matrices

Bibliographic coupling

NetMatrix <- biblioNetwork(database, analysis = "coupling", network = "authors", sep = ";")

# calculate jaccard similarity coefficient
S <- normalizeSimilarity(NetMatrix, type="jaccard")

# plot authors' similarity (first 20 authors)
net=networkPlot(S, n = 10, Title = "Authors' Coupling", type = "fruchterman", size=FALSE,remove.multiple=TRUE)

Bibliographic co-citation