Assessing the industrial opportunity of academic research with patent relatedness: A case study on polymer electrolyte fuel cells

The detection of promising academic research is vital for firms in a variety of sectors. Bibliometric tools can be used to detect such research hidden in a pile of papers and patents; however, the relationship between academic research and industrial technology development has not been well documented. In this paper, we introduced patent relatedness, which measures the semantic similarity of papers and patents, and conducted a case study on polymer electrolyte fuel cells (PEFC). The results show that in an academic research area with a small number of papers, recent average publication year, low patent relatedness has a high potential to increase in subsequent years. Research areas are identified by clustering the citation network of academic papers, and their patent relatedness and time series variation were measured and analyzed. Our results showed that potential research areas were characterized by small but emerging features. Using these findings, we identified the potential PEFC research areas and the research capability of each country.