Mapping the Semantic Web with Bibliometric Analysis: An Intellectual, Conceptual, and Social Analysis
Rohit Singh, Ravindra Kumar, Shams Khan
The study investigates the evolution and likeness evolution of scientific foundations and core research themes in three domains Semantic web, linked data and open linked data. Bibliometric study of 3021 research papers is carried out using Scopus database data for 2004-2026. Using VOSviewer, which is a software tool used for the construction and visualisation of bibliometric networks, some analyses namely, bibliographic coupling, keyword co-occurrence and the co-authorship analysis were performed to characterise the intellectual, conceptual and social structures of the field Bibliographic coupling analysis shows not only one or few major core research areas, but also many more niche smaller areas co-exist. One major cluster created by experts a Core Linked Data Principles, Knowledge Graph Embeddings and Machine Learning, Core Semantic Web Theory. According to the data from keyword co-occurrence analysis, a hierarchical conceptual model for data science, comprising three clusters, namely, fundamental theory, facilitating technologies and practical applications, has been identified. The co-authorship study, reflecting an internationalisation of researches, indicates the presence of three sizeable European and North American hubs. The primary aim of this effort is to produce a detailed multi-layer ‘science map’ of the core Semantic Web research community. The analysed information indicates substantial progress into the field. The strong international cooperation can support a dynamic evolving diverse ecosystem of applied research owing to the robust theoretical framework. Researchers can benefit greatly from this map, especially those who wish to learn about the main research fronts, the most influential documents and the general structure of the field. Moreover, it also helps to identify the larger intellectual gaps and possible direction for future further works.