Small-Data and Its Application among Various Scientific Areas: A Scientometric Study

The purpose of this study is to identify the characteristics of scientific products in the field of small-data indexed in the Web of Science database and to explain its application based on identifying the words of scientific products related to this subject separately by scientific fields.

This research is a descriptive study based on the scientometric approach and content analysis method, which has been done by using the common techniques of co-word analysis and social network analysis. Data analysis was performed by HistCite, Bibexecl, Gephi, and SPSS software; and the data mapping is done by VOSviewer.

Over the past decades, the rate of publications in the field of small data has had an increasing trend with an average annual growth rate of 15.59%. The main language of these works is in English. Although the National Cheng Kung University (Taiwan) ranked the first of organizations in this field, the United States, China and Germany recognized the top countries in this field, overall. More than 90% of these products are in the fields of Computer Science (8 clusters), Engineering (6 clusters), Mathematics (7 clusters), Telecommunications (5 clusters), and Physics (3 clusters). The greatest degree of centrality belongs to Machine Learning, the Internet of Things, and Universal existence; the most closeness centrality belongs to Adaptation, Bipartite Graph, and Machine Learning; and the most betweenness centrality belongs to Machine Learning, Long-Term Evolution Technology, and Global Existence.

The pattern of dissemination of scientific products in the field of small data indicates a continuous growth situation. Theoretical discussions of microdata have further evolved in mathematics and physics, and its applications in computer science and other fields are expanding.