**Figure 3.**

*Selecting the data to be exported on the Scopus platform [15].*

*Bibliometric Method for Mapping the State-of-the-Art and Identifying Research Gaps… DOI: http://dx.doi.org/10.5772/intechopen.85856*


**Figure 4.**

*Scientometrics Recent Advances*

Publication period

will also be returned.

*Source: Prepared by the authors.*

**Table 1.**

**52**

**Figure 3.**

*Selecting the data to be exported on the Scopus platform [15].*

One of the errors in the mining of bibliometric data is the use of a word that has various meanings. For example, if you search for the term "Lean" many articles that belong to "Lean Manufacturing" search field will be selected. However, many articles that contain the word "Lean" to refer to a characteristic of the human body

**Inclusion criteria Exclusion criteria**

Up to 2017 Articles published in 2018

Document type Articles and reviews Conference papers, book chapters, books, notes,

Source type Journals Books, websites, conference proceedings, trade

—

letters, editorials

publications, doctoral thesis, masters dissertations

Database Scopus and WoS Others databases

Subject area Physical, life, social, health

and humanities sciences

Language English Other languages

*Example of application of the search criteria for the study field "Lean Six Sigma".*

For this reason, it is essential that the author be attentive to this analysis on the search results. It is recommended that the title of the article be analyzed first to verify its suitability for the purposes of the study. If there is any doubt after the examination of the title, it is necessary to read the article abstract. If the doubt persists, it is fundamental to analyze the full text to decide if the bibliometric data

Scientific platforms provide various bibliometric data, and among the most relevant to perform the bibliometric analysis are title of the article, authors, journal, year of publication, number of citations, institutions, countries, keywords, and bibliographic references. The choice of exporting all these data should be specified

Scientific platforms offer different options of file formats to be exported. The most suitable file formats for export are ".csv" and ".txt". These formats allow the bibliometric data to be analyzed in spreadsheets and/or bibliometric softwares.

of this article will be exported to the software of bibliometric analysis.

in the scientific platform itself, as shown in **Figures 3** and **4**.

*Selecting the data to be exported on the WoS platform [16].*


### **Table 2.**

*Imported bibliometric data and bibliometric parameters generated.*

In order to make bibliometric analysis more accessible to the scientific community, the presented method recommends the use of open source software such as Calc (LibreOffice), Sci2 Tool (CNS), and VOSviewer (Leiden University).

Calc is a spreadsheet, and its use covers two main purposes: to integrate bibliometric data from different scientific platforms and analyze them. The bibliometric data imported in the Calc spreadsheet and the respective bibliometric parameters generated are presented in **Table 2**.

Importing data into the spreadsheet will only occur when the files are exported in the correct format. For the Scopus platform, the file should be exported in the ".csv" format. On the WoS platform, the data should be exported in the ".txt" format ("tab-separated" option). With the data imported into separated worksheets, the researcher must organize them into a single worksheet and classify according to the title of the article. This activity will allow the identification of duplicate files, making possible their elimination.

The bibliometric parameter "evolution of publications" is based on the publication year of each article. To obtain this parameter, the spreadsheet must be operated to quantify the number of articles published per year. This parameter reveals the dynamics of the publications, whether the topic has been widely explored by researchers or if there is a reduction in the interest of the scientific community.

The bibliometric data of research areas are used to quantify which areas are the most researched ones. The information obtained by this bibliometric parameter

allows to verify in which areas the researchers have directed their efforts and those that are still underexplored.

Keywords are concise representations of a given text that allow readers to identify its content in advance [17]. The keywords bibliometric data can be used to construct frequency charts of the most used keywords, as well as to elaborate "word clouds".

It is important that the researcher pay attention to keywords similar to terms or expressions already used in the search filtering process. Since these keywords represent obvious information regarding the topic addressed, it is recommended to disregard them to not analyze them in the subsequent analysis. This avoids inconsistencies in the analysis and wasting time and effort of the researcher.

In the case of publications that do not use keywords (something very common in administration journals, for example), the researcher must review the titles (and even the abstracts, in the case of inaccurate titles) to identify terms or expressions that define the main topics covered therein [11, 18].

The word cloud is a practical tool for dimensional visualization of the most used keywords that represent the articles. The Wordle website (www.wordle.net), for instance, provides this tool. When elaborating the word cloud, the researcher should note that expressions such as "Lean Six Sigma" can be understood as three distinct words. To correct this interpretation in the website suggested, the researcher must use the "~" symbol so that the expression will be understood appropriately.

The titles of the articles are bibliometric data used to expose the most cited articles by means of the number of citations received. A simple classification of the articles according to the number of citations can reveal which are the most relevant in the researched literature. Besides the classification of the most cited articles, it is also useful to generate a chart with the evolution of the citations received by the articles over the years. This chart can clarify which articles have increased its importance in the scientific world.

To identify the authors, institutions, journals, and countries most cited in the literature, it is recommended to apply the following steps: (1) visit the Scopus and WoS platforms (logged into your respective accounts) and check in the researcher's list of the authors, institutions, journals, and countries with a minimum number of publications; (2) select the authors, institutions, journals, and countries that meet the requirement set out in step 1; (3) count in the spreadsheet the number of citations received by authors, institutions, journals, and countries; and (4) classify the authors, institutions, journals, and countries according to the highest number of citations received.

It is important to clarify the reason for filtering a minimum number of publications from authors, institutions, journals, and countries applied in step (1). If this filter is not applied, authors who had one or two publications that had a high number of citations will be included in the ranking of the most relevant authors in the literature. This often occurs with researchers who are not specialists in the field of study. For example, one author co-authored two articles that had a high impact on the literature, but the analysis of his research profile reveals that he does not act in the field of study investigated. Therefore, the author of this example may have contributed punctually to the research, but is not a specialist in the subject study.

Tables containing the top 10 authors, institutions, journals, and countries should be generated. In addition to the number of citations received, these may contain other performance information, as presented in **Table 3**.

After obtaining the bibliometric parameters, it is recommended that the bibliometric networks of the research be generated. In this chapter, it is recommended to use VOSviewer software, which will allow the construction of bibliometric networks referring to articles, authors, journals, institutions, and countries based on

**55**

next section.

presented.

**and trends of research**

tion in international research groups, etc.

number of publications of a given author.

*Bibliometric Method for Mapping the State-of-the-Art and Identifying Research Gaps…*

number of publications, number of citations

number of citations, evolution of citations over the years

Position in the citation ranking, title of the article, authors, journal, year of publication,

Position in the citation ranking, name of the author, institution, H-index, number of publications, number of citations, evolution of citations in the field of study over the

Position in the citation ranking, institution name, country, number of publications,

Position in the citation ranking, country name, number of publications, number of

Position in the citation ranking, name of the journal, publishing area, SJR or JCR, ISSN,

citation, co-citation, co-authorship, and bibliographic coupling. The software also allows quantifying terms, a useful tool to characterize the occurrence of keywords [19]. Examples of networks will be presented and explained in more detail in the

**3. Mapping of the state of the art and identification and analysis of gaps** 

The main objective of the analysis of gaps in the literature is to provide the researcher ways to identify opportunities for exploring new relevant topics of a given field of research, as well as studies that have not been fully explored and which require the development of further studies in order to advance the state of the art of a particular theme. Therefore, in this section some types of analysis that may converge in the identification of scientific gaps and research trends will be

**3.1 Analysis of authors, countries, institutions, articles, and keywords**

The analysis of most cited authors is helpful to the researcher to know the authors who are reference in the area under study and then always be attentive to new studies published by these authors. In this analysis, it is recommended that authors with a minimum number of publications be selected, as this will avoid the selection of authors who are not experts in the field. This type of analysis may indicate to the researcher opportunities for co-orientation partnerships, participa-

The analysis of the most cited authors in a theme and their most recent articles on this topic is a way of guiding the development of a bibliometric article that aims at the identification and analysis of gaps and trends. The selection of authors with a minimum number of publications can be done manually in the main research platforms or it can be done by means of the software VOSviewer, which performs the analysis of most cited authors with the option to select the filter of minimum

The analyses of most cited countries and institutions are other possibilities of bibliometric analyses that can help the researcher to direct his academic plans,

*DOI: http://dx.doi.org/10.5772/intechopen.85856*

years

citations

*Proposal of additional data for citation ranking tables.*

number of citations

**Table Table columns**

Most cited articles

Most cited authors

Most cited institutions

Most cited journals

Most cited countries

**Table 3.**

*Source: Prepared by the authors.*

*Bibliometric Method for Mapping the State-of-the-Art and Identifying Research Gaps… DOI: http://dx.doi.org/10.5772/intechopen.85856*


### **Table 3.**

*Scientometrics Recent Advances*

that are still underexplored.

appropriately.

tance in the scientific world.

citations received.

allows to verify in which areas the researchers have directed their efforts and those

tencies in the analysis and wasting time and effort of the researcher.

that define the main topics covered therein [11, 18].

Keywords are concise representations of a given text that allow readers to identify its content in advance [17]. The keywords bibliometric data can be used to construct frequency charts of the most used keywords, as well as to elaborate "word clouds". It is important that the researcher pay attention to keywords similar to terms or expressions already used in the search filtering process. Since these keywords represent obvious information regarding the topic addressed, it is recommended to disregard them to not analyze them in the subsequent analysis. This avoids inconsis-

In the case of publications that do not use keywords (something very common in administration journals, for example), the researcher must review the titles (and even the abstracts, in the case of inaccurate titles) to identify terms or expressions

The word cloud is a practical tool for dimensional visualization of the most used keywords that represent the articles. The Wordle website (www.wordle.net), for instance, provides this tool. When elaborating the word cloud, the researcher should note that expressions such as "Lean Six Sigma" can be understood as three distinct words. To correct this interpretation in the website suggested, the researcher must use the "~" symbol so that the expression will be understood

The titles of the articles are bibliometric data used to expose the most cited articles by means of the number of citations received. A simple classification of the articles according to the number of citations can reveal which are the most relevant in the researched literature. Besides the classification of the most cited articles, it is also useful to generate a chart with the evolution of the citations received by the articles over the years. This chart can clarify which articles have increased its impor-

To identify the authors, institutions, journals, and countries most cited in the literature, it is recommended to apply the following steps: (1) visit the Scopus and WoS platforms (logged into your respective accounts) and check in the researcher's list of the authors, institutions, journals, and countries with a minimum number of publications; (2) select the authors, institutions, journals, and countries that meet the requirement set out in step 1; (3) count in the spreadsheet the number of citations received by authors, institutions, journals, and countries; and (4) classify the authors, institutions, journals, and countries according to the highest number of

It is important to clarify the reason for filtering a minimum number of publica-

After obtaining the bibliometric parameters, it is recommended that the bibliometric networks of the research be generated. In this chapter, it is recommended to use VOSviewer software, which will allow the construction of bibliometric networks referring to articles, authors, journals, institutions, and countries based on

tions from authors, institutions, journals, and countries applied in step (1). If this filter is not applied, authors who had one or two publications that had a high number of citations will be included in the ranking of the most relevant authors in the literature. This often occurs with researchers who are not specialists in the field of study. For example, one author co-authored two articles that had a high impact on the literature, but the analysis of his research profile reveals that he does not act in the field of study investigated. Therefore, the author of this example may have contributed punctually to the research, but is not a specialist in the subject study. Tables containing the top 10 authors, institutions, journals, and countries should be generated. In addition to the number of citations received, these may

contain other performance information, as presented in **Table 3**.

**54**

*Proposal of additional data for citation ranking tables.*

citation, co-citation, co-authorship, and bibliographic coupling. The software also allows quantifying terms, a useful tool to characterize the occurrence of keywords [19]. Examples of networks will be presented and explained in more detail in the next section.
