Topic modeling and technology forecasting for assessing the commercial viability of healthcare innovations

• Published in: Technological forecasting & social change Vol. 156; p. 120041
• Main Authors: Erzurumlu, S. Sinan, Pachamanova, Dessislava
• Format: Journal Article
• Language: English
• Published: New York Elsevier Inc 01.07.2020; Elsevier B.V; Elsevier Science Ltd
• Subjects: Analysis; Business schools; Commercial viability; Commercialization; Data mining; Empirical analysis; Forecasting; Forecasts and trends; Health care; Health services; Healthcare; Information management; Innovations; Inventors; Licensing (technology); Mathematical models; Modelling; Patent analysis; Science and technology; Survival analysis; Technological change; Technological forecasting; Technology assessment; Technology commercialization; Technology forecasting; Topic modeling; Viability; Topic modeling; Technology forecasting; Healthcare; Commercial viability; Patent analysis; Technology commercialization
• ISSN: 0040-1625; 1873-5509
• DOI: 10.1016/j.techfore.2020.120041

•Forecasting future technologies that impact the likelihood and pace of commercialization is a key task in advancing commercial viability•Patent-based topic modeling approach combined with experts’ judgment identifies non-predetermined technologies•Survival analysis of patent licensing measures each technology's impact on commercial viability in terms of the pace and scope of commercialization•Mapping representative themes extracted via topic modeling to collaborative patent classifications allows for assessing the scope of an invention•Less complex healthcare technologies aim for commercializing in a wider group of sectors•More complex healthcare technologies that aim for commercializing in a specific sector demonstrate commercial viability sooner Developing technologies for a transfer from science to market is a key element of research-intensive organizations such as innovation management centers that work closely with inventors to commercialize their technological innovations. To advance the commercial viability of technological innovations, this paper proposes a framework that integrates topic modeling, survival analysis, and judgment of subject matter experts to forecast promising technologies using patents as data resources. Regarding the commercial viability of identified technologies, our empirical analysis focuses on patents and licensing data from a prominent innovation management center over a 20-year period. We are able to identify technologies that are statistically significant for predicting the likelihood and the time until a patent is commercialized, and suggest a way to assess their scope of technological impact. Our results contribute to the understanding of promising healthcare technologies and help R&D managers to develop the knowledge they need to advocate technologies in support of commercial returns.