Abstract: 【 Purpose/significance】 With the rapid development of AI technology, the updating and iterative process of emerging tech⁃ nology is getting shorter and shorter, and the papers and patents related to emerging technology have become the focus of scientific re⁃ searchers′ attention, and there is an urgent need to excavate the correlations and evolutionary patterns between the topics of scientific papers and emerging patented technology. 【Method/process】 From the perspective of multi-dimensional perspective analysis, using co-occurrence network analysis, theme model, text similarity calculation, and technology life cycle division as the technical support, combining the correlation analysis method under the perspective of social network analysis and the correlation analysis method under the perspective of text mining, we have constructed a scientific paper-patent technology correlation model based on a hybrid method. 【Result/conclusion】 Taking the field of biomedical technology as an example, the algorithm model is validated and the experimental results show that the cosine similarity values between "paper-patent" topics are high in three time intervals, and the topic areas have formed a trend of synergistic innovation between scientific research and patent technology. 【Innovation/limitation】 This paper pro⁃ poses a set of systematic "scientific paper-patent technology" correlation relationship construction method system, which realizes a more fine-grained and dynamic analysis of the interaction law between science and technology, and then digs out the development di⁃ rection of technical topics with innovation potential, and provides practical paths for guiding enterprises to carry out scientific and technological innovation activities.