Classification number co-occurrence network construction method, technical opportunity identification method and system

Abstract

The invention discloses a construction method of a classification number co-occurrence network, and a technical opportunity recognition method and system, and belongs to the field of technical opportunity recognition. Semantic information and co-occurrence information of joint patent classification (CPC) are combined to form a CPC co-occurrence network, hidden connection modes among CPC nodes are mined through a graph neural network model, then the CPC nodes possibly connected with artificial nodes representing a target field are predicted, and finally technical development points possibly appearing in the target field in the future are recognized. And a certain decision support is provided for formulating a technology development strategy. According to the method, potential association between the semantic information of the CPC and the co-occurrence information mining technology can be fully combined, rich node features provide a good basis for model learning, and technical opportunity recognition can be better helped to be carried out.

Description

technical field [0001] The invention belongs to the field of technical opportunity identification, and more particularly, relates to a method for constructing a classification code co-occurrence network, a method and a system for identifying technical opportunities. Background technique [0002] Technological opportunities are to discover the latest technological trends by mining the existing technological development trends and interrelationships in a certain technological field, and infer the possible technological forms or technological development points in this field. As the basis for innovation decision-making, technological opportunity is an important factor that must be considered in any technological innovation. [0003] At present, the main methods of technology opportunity identification are: based on outlier detection, based on patent map, based on link prediction, science and technology correlation method and composite method. Among them, the core steps of the ...

AI Technical Summary

Problems solved by technology

The former uses the technical association represented by the "keyword pair" as the final prediction result, and the vague semantic relationship can easily lead to deviations in the understanding of technology development trends by R&D personnel.

The latter expresses technical opportunities in the form of "patent classification code pairs", but because the meaning of patent classification codes is more macroscopic, it is not conducive to extracting specific technical details; the method of linking science and technology mainly mines the technological evolution path of papers and patent data And conduct association analysis to find topics that are more active in scientific research but have not yet appeared in the technical field. The clustering method used only retains the main part of the data, and valuable information may be lost for fields with sparse references; compound type The method refers to the identification of technical opportunities by combining multiple analysis methods. The accuracy of identification can be improved through the complementary advantages of the methods. It is an immature direction that remains to be explored.

Images