Technological trend mining: identifying new technology opportunities using patent semantic analysis

Abstract

This study identifies technological evolution patterns based on the purpose and effect of the technology using patent data. Furthermore, the direction of future development is proposed by referencing the evolution patterns in other sectors as an approach for discovering technology opportunities for companies. In order to achieve the aim, first, patent semantic analysis is conducted for extracting the technological purpose/effect in the patent document. Second, clustering is performed on the extracted purposes/effects to create a dictionary. Third, using dictionary and sequential pattern mining, the patterns of technological purposes/effects by year of the target field patent were identified. Fourth, a reference field for predicting the technological purpose/effect of the target field is selected, and the technological purpose/effect of the target field is predicted based on the evolution pattern of the technological purpose/effect in the reference field. The case study findings indicate that technologies in the bio-healthcare industry have evolved towards enhancing data quality or energy efficiency after ensuring functional diversity. Referencing the evolutionary trends in the telehealth industry, technologies in bio-healthcare can consider improving consumer convenience as technology opportunities at the macro level, along with supporting product use, particularly in various medical conditions, to ultimately realize automation at the micro level. The proposed approach enables an in-depth understanding of technological evolution patterns via empirical analysis of patent data and ultimately supports the identification of new technology opportunities by comparing evolution patterns in the target field with the reference field.