Biblioshiny

bibliometrix for no coders

MASSIMO ARIA
FULL PROFESSOR IN STATISTICS FOR SOCIAL SCIENCES – UNIVERSITÀ DEGLI STUDI DI NAPOLI FEDERICO II
CORRADO CUCCURULLO
FULL PROFESSOR IN MANAGEMENT AND ECONOMICS – UNIVERSITÀ DEGLI STUDI DELLA CAMPANIA “LUIGI VANVITELLI”
bibliometrix: An R-Tool for Comprehensive
Science Mapping Analysis
Aria, M., & Cuccurullo, C. (2017). bibliometrix: An R-tool for comprehensive science mapping analysis. Journal of Informetrics.

bibliometrix is an open-source tool for executing a

comprehensive science mapping analysis of scientific
literature
It was programmed in R language to be flexible and
facilitate integration with other statistical and
graphical packages. Indeed, bibliometrics is a constantly
changing science and bibliometrix has the flexibility to
be quickly upgraded and integrated
Its development can address a large and active
community of developers formed by prominent
researchers
bibliometrix
bibliometrix provides various routines for importing bibliographic data
from SCOPUS, Clarivate Analytics' Web of Science, Dimensions, The
Lens, PubMed and Cochrane databases, performing bibliometric
analysis and building data matrices for co-citation, coupling, scientific
collaboration analysis and co-w
Biblioshiny App

Biblioshiny is a web-base app included in the bibliometrix package

Biblioshiny allows no coders to use bibliometrix

It is developed in the Shiny environment

Just install and load bibliometrix, library(bibliometrix) and type

biblioshiny() and the game starts!
 Biblioshiny: how it works
Biblioshiny combines the functionality of bibliometrix package with the
ease of use of web apps using the Shiny package environment
Shiny R environment
biblioshiny web app bibliometrix
dashboard
package package

What do you see! What it do!

 The Data
What, Where, How
Bibliographic database
A bibliographic database is a database of bibliographic records, an
organized digital collection of references to published scientific
literature, including journal articles, conference proceedings,
patents, books, etc.

They generally contain very rich subject descriptions in the form of

keywords, subject classification terms, or abstracts.

Information related to a bibliographic record are named

bibliographic meta-data
Main bibliographic databases
Multidisciplinary:
Microsoft Academic
CrossRef
Dimensions
OpenAlexR
Web of Science
Scopus
…
Specialized:
ArXiv
Cochrane
EconBiz
IEEE Xplore
PubMed
…
(Source: Visser, M., van Eck, N. J., & Waltman, L. (2021). Large-scale comparison of bibliographic data sources: Scopus,
Web of Science, Dimensions, Crossref, and Microsoft Academic. Quantitative Science Studies, 2(1), 20-41.
An example about document meta-data

Meta-Data
Dataset (analyzed by biblioshiny)
Supported databases: File formats:

Web of Science (WoS) Plain text

Scopus BibTeX
Dimensions CSV/xlsx
Lens.org CIW
PubMed zip (Multiple files importing)
Cochrane Rdata (bibliometrix file)
Openalex
 Bibliographic record of: Aria & Cuccurullo, 2017, Journal of Informetrics
An example of WoS
plain text export format
Main meta-data fields:
AU Authors
AF Authors’ full name
TI Title
SO Document source (eq Journal name)
DT Document type
DE Authors’ keywords
… …
ID Keyword Plus (assigned by WoS machine learning algorithm)
AB Abstract
C1 Authors’ affiliations
RP Corresponding author’ affiliation
CR Cited references
TC Total citations
PY Publication year
DI DOI
SC Subject category
…
For a complete list of field tags see
http://www.bibliometrix.org/documents/Field_Tags_bibliometrix.pdf
An example of BibTeX export formats
Web of Science Scopus

Very different!
(both in terms of content and string format)

The merging of WoS and Scopus collections

is a very difficult task
(no software currently allows this!)
Some remarks about the DBs and the data
formats
Web of Science is preferable to other databases in terms of data
quality
In Scopus, the reference elements mentioned are not standardized -> they must
be combined
In Dimensions, the algorithm that classifies search areas is not efficient

In Web of Science, plain text format is preferable to others

Scopus BibTeX format and Dimensions CSV format do not allow exporting some
metadata
Some limits of the DBs
Web of Science:
Permits to export only 500 records at a time but allows to split the selected collection into multiple
downloads (e.g. from 1 to 500, 501 to 1000, 1001 to 1500, …., from 5001 to 5500, etc.)
It is not possible to use API to directly search and export meta-data (using the Italian academic subscription)
Scopus:
Scopus permits to export 2,000 records at a time but does not allow to split the selected collection into
multiple downloads (It is necessary to define a multiple search strategy selecting up to 2,000 documents at a
time!)
It allows scholars to use API to directly search and export meta-data in R environment (e.g. rscopus package)
Dimensions:
Dimensions allows you to export 50,000 records at a time but does not allow you to split the selected
collection into multiple downloads (It is necessary to define a multiple search strategy by selecting up to
50,000 documents at a time!)
It allows scholars to use the API to search and export metadata in R environment (e.g. dimensionsR package)
 Data Collection
Querying, Selecting, Exporting
“Querying” a bibliographic database
Data can be extracted through a query
A query is a combination of terms linked by Boolean operators
A query defines a search strategy by (search fields):
Keywords
Titles
Abstracts
Authors
Journals
Affiliations
…
An example using the Web of Science database

“The use of bibliometric approaches in information science and library

science disciplines”

We want to describe the use of the bibliometric approaches in the

information science and library science scientific literature
Let’s start with the example:
How to define a query?
1) We need to choose the combination of terms which identify the
scientific literature that used bibliometric approaches
2) We need to limit research in the area of information science and
library science disciplines
3) We need to limit the timespan
4) We need to choose what kind of documents to analyze
PRISMA diagram
PRISMA stands for Preferred Reporting Items for
Systematic Reviews and Meta-Analyses
It is an evidence-based minimum set of items for
reporting in systematic reviews and meta-
analyses
The aim of the PRISMA Statement is to help
authors improve the reporting of systematic
reviews and meta-analyses
PRISMA may also be useful for critical appraisal of
published systematic reviews, although it is not a
quality assessment instrument to gauge the
quality of a systematic review
Search strategy

1. Select the WoS sub-DBs: SCI (Science Citation Index), SSCI (Social
Science Citation Index), and ESCI(Emerging Sources Citation
Index
2. Select all the documents that contain the words “bibliometric*”
or “science map*” in the title, abstract or in the keyword list
3. Select only documents included in the subject category
Information Science Library Science
4. Select the timespan: all complete years
5. Select document types: Articles or Proceedings Papers or Review
Articles
6. Select only documents written in English
 Search strategy − Step 1

Select the WoS sub-DBs:

SCI - EXP
EXPA
AND
NDED ED (Scie
(Scienc
nc
ncee Cita
itattion Inde
Indexx)
SSC
SSCII (So
(Socia
cia
ciall Scie
Scienc
nc
nce
e Cita
itattion Inde
Indexx)

ESCI (Emer
(Emergging Sour
Sources
ces Cit
Citat
at
ation
ion Inde
Index)
x)
 Search strategy − Step 2

Select all the documents that contain the words

“bibliometric*” or “science map*” in the topic
“*” is a jolly
symbol
it means any
character Search field:
“title”
e.g. bibliometric*: “topic”
- Bibliometric “keywords”
- Bibliometrics …
- Etc.
e.g. science map*: Boolean operator:
- Science map “OR”
- Science maps “AND”
- Science mapping “NOT”
 Search strategy − Step 2

Results for "bibliometric*" (Topic) OR "science map*"

(Topic)

Query result:
20,248 documents
 Search strategy − Step 3
WoS categories:
Information Science Library Science

Refine by
WoS category:
Information
Science Library
Science

Results:
5,390
documents
 Search strategy − Step 4

NOT Publication Years: 2022

Exclude incomplete years: 2022

Results: 5,329 documents
 Search strategy − Step 5-6
Document types:
Articles or Proceedings
Papers or Review Articles Languages: English

Refine by document type:

articles, proceedings papers, and Refine by language: English
Results: 4,441 documents
review articles
Results: 5,107 documents
 Search strategy − Marked list

Continue the search or export the results

Now we should continue our search strategy (using other filters) or export the results
 Search strategy − Search history

Search history

History represents
a classical
PRISMA diagram
It shows the
search steps we Clicking on a
just performed result, we can go
back along the
search steps.
 Search strategy
Difference between
“Add to marked list” and “export”
Export allows you to immediately export the results of your search

Marked List is a sort of “shopping cart” where we can save your

meta-data collection and we can continue to manipulate or export
the results of your search in the future
 Search strategy − Marked list

“Add to marked list”

 Search strategy − File export

“Export”
Repeat this operation until the total
number of documents is reached (e.g.,
from 1 to 500, from 501 to 1000, etc.)

WoS will download many plain text files

called: saverecs.txt, saverecs (1).txt, …
Biblioshiny interface
Tabs, Methods, Workflow
bibliometrix package installation

Le
Lett’s start pla
playy with bib
biblios
lios
liosh
hin
iny!!
y!!
y!!!!
 Biblioshiny interface
Notifications about
software

-Package tutorial
-Information about
Convert and import
data
-A biblioshiny tutorial
Possibility of making
donations that help
After using the ensure the future
software development of
Bibliometrix.
remember to
Link to:
cite bibliometrix
K-Synth
Github
 Welcome Tab

Biblioshiny workflow

4 Levels of
analysis
3 Structures
of Knowledge

Organized according to the science mapping workflow

 Bibliometric Analysis for Systematic Literature
Reviews
Bibliometric Analysis Often, new knowledge
emerges at crossroad
Focus on domain Clustering by Coupling K structures among structures and
(metrics) time evolution
Level of Analysis Metrics Bibliometric Unit of Statistical
Structure
Overview Main information Technique Analysis Techniques
Annual scientific production Conceptual  Co-word  ID, DE (keywords)  Network Analysis
Average citations per year  TI  Thematic mapping
Three-Field Plot  AB  Thematic evolution
Sources Most relevant and cited  Subject categories (Wos)  Factorial Analysis (CA; MCA; MDS)
Bradford’s Law Intellectual  Co-citation  Papers  Network analysis
Impact metrics  Citation  Authors  Histograph
Source dynamics  Sources
Authors Most relevant and cited Social  Collaboration  Authors (co-authorship)  Collaboration network
Production over time  Institution
Affiliations
Lotka’s Law  Countries
Countries
Impact metrics
Documents Most local/global cited
Cited references Spectroscopy
Words: ID, DE, TI, AB Words: word cloud, treemap
Trend topic
 Data analysis
Loading, Converting, Filtering
 Load Tab

Loading and converting a meta-data collection

biblioshiny allows you to

Import or Load files or
Gather data using the APIs
(Application Programming
Interfaces)
 Load Tab

Loading and converting a meta-data collection

biblioshiny can
Database
download data via
selection
APIs

bibliometrix provides
support for the APIs Number of
Documents
Data downloaded through this of Dimensions, NCBI
API cannot be used for PubMed and Scopus.
identifying citation,
bibliographic coupling, and co- Export meta-
citation links between items. data
 Load Tab

Loading and converting a meta-data collection

By default, the access
to PubMed API is free
and does not
necessarily require an
API key. In this case,
PubMed limits users to
making only 3
requests per second.
Users who register for
an API key are able to
make up to ten
requests per second
 Load Tab

Loading and converting a meta-data collection

Dimensions API needs
an account to obtain a
valid token to query the
database.
The account can be
obtained for free for
scientometric research
project asking for it at
https://www.dimensions
.ai/scientometric-
research/
 Load Tab

Loading and converting a meta-data collection

biblioshiny allows you to

import raw files
(.bib, .txt, .ciw, .csv),
load a bibliometrix file(s)
(.rdata, .xlsx),
or use a sample collection,
from from the major
bibliometric databases.
 Load Tab

Loading and converting a meta-data collection

Database
selection
File selection
Number of
Documents
Convert/Export
meta-data
Multiple export files (e.g. savedrecs.txt, savedrecs(1).txt., etc) can be imported by a single “.zip” file
 Load Tab

Loading and converting a meta-data collection

Bibliographic dataframe:
• Each row is a document
• Each column is a meta-data field
 Filter Tab

Filtering data

Language (LA)
Publication year (PY)
Document Type (DT)
Average Citation per Year
Source by Bradford’s Law

In our example, we do not need to apply any filter