Identifying and Assessing Innovation Pathways for Emerging Technologies: A Hybrid Approach Based on Text Mining and Altmetrics

Abstract

Accurately identifying core technological innovation pathways (TIPs) and evaluating opportunities in emerging technologies is important. In this article, we present an integrated framework that combines sentiment analysis, subject-action-object (SAO) analysis, machine learning, altmetrics, and expert judgments to extract technical intelligence, chart technological innovation pathways, and analyze which research avenues have the most future promise. For industry stakeholders, our approach provides a comprehensive, time-efficient, and future-oriented evaluation system to support decision making. For researchers, our methodology should inspire new ways of thinking about technological opportunity analysis—particularly, exploring the merits of finding new and advantageous combinations of existing bibliometric and nonbibliometric techniques, rather than reinventing the wheel. For example, this methodology is to integrate bibliometric analysis with sentiment analysis to gauge the attitudes of domain experts toward a topic's potential value. We combine SAO analysis and machine learning to identify relationships between topics from tremendous short texts. Beyond traditional techniques, we have also drawn on altmetrics to further validate the findings our of analysis. A case study on gold nanoparticles demonstrates the merits of our framework, revealing anti-cancer therapies and dye-sensitized solar cells (DSSCs) as the two applications with most future potential and societal impact in this field.