Patent knowledge graph construction method and device

Abstract

The invention discloses a patent knowledge graph construction method and device, and the method comprises the following steps: crawling semi-structured data of a patent related website from the Internet; processing the semi-structured data to obtain initial patent data; obtaining the patent attribute of each patent according to the initial patent data, and constructing a patent knowledge graph according to the IPC classification tree corresponding to the IPC condition of the patent attribute of each patent. Knowledge graphs are created by utilizing the advantages of IPC classification numbers.Knowledge is extracted from titles and abstracts for body and knowledge extension, so that not only is theme refinement facilitated, but also steps of manual labeling can be reduced, extension of theknowledge graph can be completed only through manual examination, and more subdivided knowledge can be obtained.

Description

technical field [0001] The present invention relates to the technical field of knowledge graph construction, in particular to a method and device for constructing a patent knowledge graph. Background technique [0002] With the development of the Internet, the content of network data presents an explosive growth trend. Due to the characteristics of large-scale, heterogeneous and diverse, and loose organizational structure of Internet content, it poses challenges for people to effectively obtain information and knowledge. Knowledge Graph, with its powerful semantic processing capabilities and open organization capabilities, has laid the foundation for knowledge-based organizations and intelligent applications in the Internet age. The knowledge map was officially proposed by [Google] on May 17, 2012. Its original intention is to improve the capabilities of search engines, improve users' search quality and search experience. With the development and application of artificial ...

AI Technical Summary

Problems solved by technology

Method (1) can construct patent databases of different themes, and constructing knowledge graphs based on topic features can be applied to many fields such as recommendation and retrieval, but directly through word segmentation of patent data, keywords are extracted and clustered topics have keywords Miscellaneous, the subject is not clear, which makes the patent types under the same subject database relatively mixed.

Method (2) can only use the existing domain and business characteristics for ontology construction, and has the disadvantage of being unable to expand other necessary ontology structures, resulting in imperfect graph construction

[0006] In summary, the current knowledge map construction is top-down map construction based on domain characteristics and business characteristics, which has the disadvantage of being unable to flexibly expand the ontology structure; there are also methods for analyzing and processing data and abstracting the ontology layer structure through various methods , there are disadvantages such as imperfect ontology construction and chaotic patent data storage under the same ontology

Images