Identifying the Trends of Technological Convergence Between Domains Using a Heterogeneous Graph Perspective: A Case Study of the Graphene Industry

Technological convergence can lead to the emergence of new products and disruptive technologies within industries, offering both opportunities and challenges. Therefore, it is crucial for enterprises to timely recognize the trends of technology convergence to make informed business decisions. In this study, we propose a deep learning method within a heterogeneous graph framework to identify technological convergence trends between domains. The approach involves collecting patent texts and metadata, constructing a heterogeneous graph network based on IPC patent, and learning representations of IPC nodes based on the IPCvec model. Using the graphene industry as a case study, we validate the effectiveness of our method by comparing it with other graph neural network models. Moreover, we estimate the trends of technological convergence across different domains in the industry at a broader data level and confirm the presence of integration over time. This method can assist enterprises in identifying technological convergence trends, uncovering new business opportunities, fostering cross-domain collaboration and innovation, guiding strategic decision-making and resource allocation, and promoting technological and industrial development. Ultimately, it enables enterprises to drive innovation, achieve growth, and enhance their competitiveness.