Seven clusters of data visualization articles in Scopus using social network analysis

,


Introduction
This article discusses the discourse Data Visualization.The discourse understanding is inseparable from bibliometric analysis (Lee, 2020;Mifrah, 2020;Omoregbe et al., 2020;Saravanan & Dominic, 2014), referring to the incorporation of various frameworks and methods to analyze citations from scientific publications.Such attempts lead to the development of different metrics to gain insight into the intellectual structure of a broad academic discipline and to evaluate the impact of a particular field of study (Akhavan et al., 2016;Putera et al., 2020).Although some researchers have produced Data Visualization articles, we have not found research in bibliometric articles on Data Visualization utilized social network analysis.The aim of this article was to provide useful data for understanding global publication trends regarding Data Visualization.This study aimed to analyze the bibliographic characteristics and trends of articles on Data Visualization published in journals indexed in Scopus written by researchers from throughout the world and to conduct an analysis of keyword co-occurrence using VOSviewer.

Methods
This study did not involve human subjects; therefore, neither institutional review board approval nor informed consent was needed.This study was a descriptive and bibliometric analysis based on a literature database.The data in this study were retrieved from the Scopus database.To obtain the necessary data, this study used the keyword "Data Visualization" in the title, abstracts, and author's keywords.In this step, we found 862 articles.In the next step, we downloaded the articles from the Scopus database and analyzed the 862 articles that had been sorted by relevance.In this study, the metadata and refined Scopus result values were retrieved in the Csv dataset format.However, before the bibliometric analysis, the consistency and reliability of the data were checked to address issues such as a lack of consistency in country names and keywords.The data were also standardized to ensure consistency regarding key words that sometimes appeared in singular or plural, abbreviations, or other forms.The data obtained from the Scopus database were analyzed using VOSviewer software, and simple statistics were calculated using Microsoft Excel.

Results
The development of 'Data Visualization' publications indicates an increase, from 2019 to 2020.The trend analysis (see Fig. 1) indicates that before 1991, there were no 'Data Visualization' publications on the Scopus database.Based on a search with the keyword "Data Visualization", the result showed approximately 862 documents.Most articles were listed under Computer Science (n=466, 30%), Engineering (n=175, 11%), Mathematics (n=150, 10%), Social Sciences (n=108, 7%), and Medicine (n=98, 6%).The full distribution of Data Visualization articles across subject areas is shown in Fig. 2. According to VOSViewer, the articles were published in 160 different journals.The highest number of articles were published in Lecture Notes In Computer Science, with 32 publications, followed by IEEE Transactions on Visualization and Computer Graphics (n=29), Journal of Physics Conference Series (n=22), Bioinformatics (n=17), and Computer Graphics Forum (n=12).The other most productive journals with the most publications are shown in Table 1.Table 1 The most production journals based on the number of publications    There were four journals from the United Kingdom, three journals from the United States, and one journal each from Switzerland, Portugal, and Germany.In the period 2020 to 2022, the United States was the country with the most publications on Data Visualization, with 305 articles, followed by China with 98 articles.China, India, Japan, Malaysia, South Korea and Hongkong were the Asian countries ranked in the top 20 countries in terms of the most Data Visualization publications.These Six Asian countries ranked two, nine, ten, sixteen, eighteen and nineteen.The top 20 countries can be seen in Fig. 3. Fig. 4 shows Corresponding Authors of Pandemic Covid-19.The highest number of articles were in the United States, Followed by China, and the United Kingdom.A content analysis was performed of the 862 publications sorted by relevance.Next, we performed a co-occurrence analysis with VOSviewer, using the "all keyword" analysis unit and the "full counting" method.We limited the frequency of keyword occurrence to 5 times; out of 6179 keywords VOSviewer found 318 keywords that met the threshold.The results of this analysis are presented in Fig. 5. Data Visualization (593), Visualization (459), and Human (136) were the top three keywords that appeared most frequently.Moreover, we found seven clusters in this analysis.Fig. 6 shows an overlay visualization of Data Visualization literature with the average number of publications from 2012 to 2020.There was a shift in topics; around 2014, the literature on Data Visualization contained extensive discussions of the terms "Information Visualization", "Information Systems" and "Data Mining", and then the last 3 years discussed "Data Analytics", "Open Source", and "Covid-19".

Discussion
Based on data from Scopus, the publication trends, journal performance, content analysis, and bibliographic coupling of countries and sources were analysed for research on Data Visualization issues throughout the world.The current study focused on articles published in Data Visualization.This study aimed to provide information on the status of publications in these fields.A total of 862 studies published were recorded in the Scopus database.The data showed the rapidity of article publications and the responsiveness of researchers in analyzing Data Visualization around the world.However, limited research from a global perspective on Data Visualization in the past 3 years has discussed "Data Analytics", "Open Source", and "Covid-19" and its relationship with governance within the scope of social science.Based on Fig. 3, the most productive and influential country was the United States followed by China.Although countries from Europe dominated the top 20 countries with the most publications by affiliated researchers, China, India, Japan, Malaysia, South Korea and Hongkong is the country from Asia in the top 20.The current study has limitations, we only retrieved studies from Scopus and did not use other source such as Web of Science, Crossref, or PubMed Central.Finally, we did not use other analyses in VOSviewer, such as co-citation or co-authorship.Thus, we hope that bibliometric research on this topic will expand in terms of the databases used, the subject areas, and the analyses conducted in order to provide a broader overview of the issue.

Conclusion
In the past 30 years, global research on Data Visualization has increased significantly.The theme of research on Data Visualization related to local government could be interesting for future discussions.There are also opportunities to foster discussion about Data Visualization in social science journals related to public administration.Finally, Europe dominated this field in terms of publications, while research from Asia on this topic remains limited, and further research is therefore necessary.

Fig. 3 .Fig
Fig. 3. Top 20 countries with publication of Data Visualization Fig. 5 shows these keywords divided into 7 clusters (each with a different number of keywords), which are represented by colors.The first cluster (red, 107 keywords) focused on Data Visualization, Visualization, Information Systems, Big Data and Decision Making.The second cluster (green, 55 keywords) centered on Software, Computer Graphics, Procedures, Priority Journal, and Data Processing.The third cluster (blue, 55 keywords) related to Data Mining, Algorithms, Data Reduction, Data Structures, and Mapping.

Fig. 5 .Fig. 6 .
Fig. 5. Network visualization of data visualization articles published globally in the Scopus database (Source: Processed by Author) The fourth cluster (yellow, 51 keywords) related to Human, Article, Methodology, Factual Database, and Female.The five clusters (purple, 28 keywords) focused on Data Analysis, Quality Control, Quantitative Analysis, Data Sharing, and Automation.The sixth cluster (light blue, 20 keywords) dealt with Three Dimensional, Virtual Reality, Students, Education, and Tools.The seventh cluster (orange, 2 keywords) focused on Interactive Computer Graphics and Research.

Table 2
The most influential authors and source based on citation analysis

Table 3
The most influential countries and source based on citation analysis.