Purpose： The purpose of this study is to modernize previous work on science overlay maps by updating the underlying citation matrix, generating new clusters of scientific disciplines, enhancing visualizations, and pr...Purpose： The purpose of this study is to modernize previous work on science overlay maps by updating the underlying citation matrix, generating new clusters of scientific disciplines, enhancing visualizations, and providing more accessible means for analysts to generate their own maps Design/methodology/approach： We use the combined set of 2015 Journal Citation Reports for the Science Citation Index （n of journals = 8,778） and the Social Sciences Citation Index （n = 3,212） for a total of 11,365 journals. The set of Web of Science Categories in the Science Citation Index and the Social Sciences Citation Index increased from 224 in 2010 to 227 in 2015. Using dedicated software, a matrix of 227 × 227 cells is generated on the basis of whole-number citation counting. We normalize this matrix using the cosine function. We first develop the citing-side, cosine-normalized map using 2015 data and VOSviewer visualization with default parameter values. A routine for making overlays on the basis of the map （＂wc 15.exe＂） is available at http：//www.leydesdorff.net/wc 15/index.htm. Findings： Findings appear in the form of visuals throughout the manuscript. In Figures 1 9 we provide basemaps of science and science overlay maps for a number of companies, universities, and technologies. Research limitations： As Web of Science Categories change and/or are updated so is the need to update the routine we provide. Also, to apply the routine we provide users need access to the Web of Science. Practical implications： Visualization of science overlay maps is now more accurate and true to the 2015 Journal Citation Reports than was the case with the previous version of the routine advanced in our paper.Originality/value： The routine we advance allows users to visualize science overlay maps in VOSviewer using data from more recent Journal Citation Reports.展开更多
Big data is one of the current and future research frontiers.It has received international attention,and some countries have even upgraded big data research to a national strategy.Therefore,it is interesting to unders...Big data is one of the current and future research frontiers.It has received international attention,and some countries have even upgraded big data research to a national strategy.Therefore,it is interesting to understand the status quo of big data research and identify the status and contribution of a country.Our study is divided into two parts.The first part of this study combines core lexical query and expanded lexical query to get relatively integral publications’data sets on big data.Citation relationships and a maximum connected subgraph algorithm are used to clean and filter unrelated publications.Then the Leiden algorithm is selected to cluster the citation network for big data and VOSviewer is used to map the big data knowledge structure.In the second part of this study,we analyze China’s research contribution in terms of research output and highly-cited papers.In order to better show the distribution of big data research in China,we utilized science overlay mapping to visualize the status quo of China’s research in big data.Our study shows that China is one of the most important countries in big data research and the research covers almost all areas of big data.However,the research performance is relatively low.In terms of knowledge structure with science overlay mapping,China’s research mainly focuses on cloud computing,the Internet of Things(Io T),and social media.However,research topics with a greater rate of highly-cited papers are mainly found in cloud computing,big data medicine,and Industry 4.0.These topics are also the dominant areas of China’s big data research.展开更多
文摘Purpose： The purpose of this study is to modernize previous work on science overlay maps by updating the underlying citation matrix, generating new clusters of scientific disciplines, enhancing visualizations, and providing more accessible means for analysts to generate their own maps Design/methodology/approach： We use the combined set of 2015 Journal Citation Reports for the Science Citation Index （n of journals = 8,778） and the Social Sciences Citation Index （n = 3,212） for a total of 11,365 journals. The set of Web of Science Categories in the Science Citation Index and the Social Sciences Citation Index increased from 224 in 2010 to 227 in 2015. Using dedicated software, a matrix of 227 × 227 cells is generated on the basis of whole-number citation counting. We normalize this matrix using the cosine function. We first develop the citing-side, cosine-normalized map using 2015 data and VOSviewer visualization with default parameter values. A routine for making overlays on the basis of the map （＂wc 15.exe＂） is available at http：//www.leydesdorff.net/wc 15/index.htm. Findings： Findings appear in the form of visuals throughout the manuscript. In Figures 1 9 we provide basemaps of science and science overlay maps for a number of companies, universities, and technologies. Research limitations： As Web of Science Categories change and/or are updated so is the need to update the routine we provide. Also, to apply the routine we provide users need access to the Web of Science. Practical implications： Visualization of science overlay maps is now more accurate and true to the 2015 Journal Citation Reports than was the case with the previous version of the routine advanced in our paper.Originality/value： The routine we advance allows users to visualize science overlay maps in VOSviewer using data from more recent Journal Citation Reports.
基金supported by the National Social Science Foundation of China(Grant No.19ZDA348)
文摘Big data is one of the current and future research frontiers.It has received international attention,and some countries have even upgraded big data research to a national strategy.Therefore,it is interesting to understand the status quo of big data research and identify the status and contribution of a country.Our study is divided into two parts.The first part of this study combines core lexical query and expanded lexical query to get relatively integral publications’data sets on big data.Citation relationships and a maximum connected subgraph algorithm are used to clean and filter unrelated publications.Then the Leiden algorithm is selected to cluster the citation network for big data and VOSviewer is used to map the big data knowledge structure.In the second part of this study,we analyze China’s research contribution in terms of research output and highly-cited papers.In order to better show the distribution of big data research in China,we utilized science overlay mapping to visualize the status quo of China’s research in big data.Our study shows that China is one of the most important countries in big data research and the research covers almost all areas of big data.However,the research performance is relatively low.In terms of knowledge structure with science overlay mapping,China’s research mainly focuses on cloud computing,the Internet of Things(Io T),and social media.However,research topics with a greater rate of highly-cited papers are mainly found in cloud computing,big data medicine,and Industry 4.0.These topics are also the dominant areas of China’s big data research.
