This paper proposes an efficient clustering method of technology fields for future technical trend prediction from the public information. The speed of product development is steadily improving due to the spreading of ICT as social infrastructure and the rapid progress of machine learning. However, in the process of finding a new solution to the problem, the developer’s capability is still an important. Referring to the problems and solutions to other technical fields with similar technology structure is one of the effective ways to find new solutions. However, selection of comparative fields often depends on the technical preferences and experience of developers. Thus, important signals might be overlooked. In this paper, we focus on the classification codes of patent for extracting the technology structure from the patent data. The link mining method is employed for visualizing the structure and extracting the feature. The structure is visualized as the graph of classification codes, and the feature is extracted as the features of the graph. From the result of the proposed method, we succeeded to reveal the cluster of the technology fields with similar technology structure.
CITATION STYLE
Shibata, M., & Takahashi, M. (2018). A method of knowledge extraction for response to rapid technological change with link mining. In Smart Innovation, Systems and Technologies (Vol. 96, pp. 299–310). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-319-92031-3_30
Mendeley helps you to discover research relevant for your work.