A digital analysis system of patents integrating natural language processing and machine learning

Technological upgradation is the driving force of social and economic development. Although technological upgradation is extensively needed, it is characterised as high intelligence, high investment, and high risk. In the face of complex market competition patterns, industrial enterprises urgently need to seek external cooperation and achieve sustainable competitive advantage. However, the current digital systems of patents fail to fulfil efficiency improvement from technology transfer to research and development (R&D) cooperation for industrial enterprise. Nor does it provide adequate support for industrial enterprises' technological upgradation. Therefore, this research proposes a systematic framework that combines nautral language processing (NLP) and machine learning (ML) technologies, including the patent recommendation model, patent transferability evaluation model, and research team detection model. Such a digital system of patents improves the efficiency of technology transactions and R&D cooperation with research institutions for industrial enterprises to identify related authorised patents and locate technology research teams. It is also expected that the developed system will enable research institutions to recommend valuable patents and transfer innovative technologies effectively.