A bibliometric study on recent trends in artificial intelligence-based suspicious activity recognition

Recent years have seen a dramatic increase in the use of artificial intelligence (AI) in suspicious activity recognition (SAR). To better understand the research work and recent trends in AI-based SAR, the paper carries out a bibliometric study to analyze the publications based on the recent developments and contributions of authors, publication source, country, and institutions, identifying the most productive items, and the partnership among each. The search on the Scopus database retrieved 1713 documents related to AI-based SAR. In this study, all document types from Scopus were included in the analysis. VOSviewer was used to perform coupling, cluster, and co-citation network analysis to identify research hotspots, while bibliometrix was used to generate keyword analysis, including word clouds, word dynamics, theme trends, and Sankey diagrams, to understand the evolution and future direction of the research field. This paper contributes valuable insights for researchers and audiences worldwide regarding emerging research areas.