Patent classification method and system based on network representation learning and hierarchical label embedding

A technology of hierarchical labeling and patent classification, applied in neural learning methods, biological neural network models, text database clustering/classification, etc., can solve problems such as low patent classification accuracy, and achieve low accuracy and improve accuracy. , the effect of promoting the effect

Pending Publication Date: 2022-05-13
ANHUI UNIVERSITY
View PDF1 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The technical problem to be solved by the present invention is how to solve the technical problem of low patent classification accuracy caused by ignoring the feature enhancement effect of the patent category label in the classification process

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Patent classification method and system based on network representation learning and hierarchical label embedding
  • Patent classification method and system based on network representation learning and hierarchical label embedding
  • Patent classification method and system based on network representation learning and hierarchical label embedding

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0074] An embodiment of the present invention provides a patent classification method based on network representation learning and hierarchical label embedding, such as figure 1 As shown, it mainly includes:

[0075] S1: Obtain patent information including patent texts, inventors, obligees, citations, and category labels; use patent citations, inventors, and obligee information to construct inventor networks and obligee networks; among them, the inventor network includes patents The relationship between nodes and between patent nodes and inventor nodes, the obligee network includes the relationship between patent nodes and between patent nodes and obligee nodes;

[0076] S2: Obtain the semantic feature representation of label description through hierarchical label embedding, and enhance the label semantics by combining the correlation between different levels in the hierarchical category structure to obtain the final hierarchical label semantic feature representation;

[00...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention provides a patent classification method and system based on network representation learning and hierarchical label embedding. The method comprises the steps that patent information containing data such as category labels is acquired; constructing an inventor network and an obligee network by using the patent quotation and the inventor and obligee information; semantic feature representation of tag description is obtained through hierarchical tag embedding, tag semantics are enhanced in combination with hierarchical category structure relevance, and final hierarchical tag semantic feature representation is obtained; patent text feature representation is obtained through text embedding, and final patent text features are obtained; network patent node features are obtained through network representation learning, patent node feature representation is enhanced in combination with the correlation between the two networks, the patent node features of the two networks are fused, and final patent node features are obtained; patent text features and patent node features are fused for patent classification. The technical problem that patent classification accuracy is low due to the fact that the feature enhancement effect of patent category labels in the classification process is ignored is solved.

Description

technical field [0001] The invention belongs to the technical field of patent classification, and relates to a patent classification method and system based on network representation learning and hierarchical label embedding. Background technique [0002] Intellectual property rights are one of the issues that countries around the world generally attach importance to, and companies usually use patents as an effective way to protect their intellectual property rights and market competitiveness. More and more technology companies use the large amount of technical information in patent databases to enhance their R&D activities, such as new product development, technology transfer, technology forecasting, and analysis of technology mergers and acquisitions. With the rapid development of various technical fields, the number of global patents has increased dramatically. As of 2020, there will be 17 million valid patents worldwide, and 3.3 million patent applications worldwide, pr...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06F16/35G06F40/30G06K9/62G06Q50/18G06N3/04G06N3/08
CPCG06F16/355G06F40/30G06Q50/184G06N3/08G06N3/044G06F18/241G06F18/253G06F18/214
Inventor 赵姝鲁凡陈嘉琳陈洁段震钱付兰张宇航
Owner ANHUI UNIVERSITY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap