Identifying convergence fields and technologies for industrial safety: LDA-based network analysis

Publication date: Available online 23 August 2018Source: Technological Forecasting and Social ChangeAuthor(s): Bomi Song, Yongyoon SuhAbstractAs industrial systems expand and complex systems are developed, it is no longer effective to minimize hazards and risks for industrial safety using the technological solutions limited to a single industry. Thus, to resolve complicated problems, safety technology has been developed by promoting technology innovation and convergence. In this respect, this study aims at monitoring major safety fields and technologies through patent analysis to identify the trends in technology development that prevent the risks of various industrial systems. Patent information is effectively used for analyzing technology descriptions, which include the purpose and newness of technology. Using this patent information, we propose the major safety fields and related technology keywords using the following two techniques: (1) latent Dirichlet allocation (LDA), which aims to extract the latent topics and main keywords contained in documents, and (2) network analysis, which is useful for monitoring change patterns and relations. Further, the convergence trajectories of safety technology are identified to provide insights about the technology trends in safety fields. The results are expected to enable safety managers and engineers to effectively find relevant technology trends for reducing hazardous factors.

Source: Technological Forecasting and Social Change - August 23, 2018 Category: Science Source Type: research