| The study focuses on the intellectual structure of linked data field by using co-word analysis and social network analysis.For this research,the data were retrieved from Web of Science(WOS)core collection database of linked data of Web of Science,including 2332 keywords from 946 articles.Using those keywords,this research investigated about 30 high frequency keywords to examine the co-occurrence matrix.Co-word analysis utilizes co-occurrence matrix for calculating factor analysis with principal component analysis,cluster analysis by Ward's method,similarity of proximity matrix of Pearson Correlation Coefficient and multidimensional scaling(MDS)with the aids of SPSS 23.Moreover,through co-occurrence matrix,social network analysis discovers the measure of centrality and k-core analysis.Therefore,the paper discovers that co-word analysis can investigate subject clusters in linked data,and can compare between subjects and intellectual structure by dividing into nine clusters which are "Internet of things",Entity linking","Education","Semantic Web","linked data","Web of data","Dbpedia","Data integration" and "Ontology".On the other hand,the social network analysis helps to describe how each keyword connects in the network,which one is influenced and how much closeness they are to each keyword.According to the results,both analysis can make and understand more about linked data field and how those fields linked each other within the subject.It can also contribute the distribution of different subject ideas in the domain or the related subjects.This paper also examines to measure the flows of relationship of each keywords and visualize the structure of relationship network.Briefly,this research improves the relative effectiveness development of using linked data in different years of articles and it aims to know how the keywords focus in the linked data field.Furthermore,co-word analysis and social network analysis are used to examine the relationships between domains and intellectuals. |
PDF Full Text Request