Chinese patent classification method and system for TRIZ invention principle and storage medium

A patent classification and patent technology, applied in the Chinese patent classification method, system and storage medium oriented to TRIZ invention principles, can solve the problems of time-consuming and labor-intensive, no unified, authoritative research methods and research results, etc., to achieve effective extraction Effect

Inactive Publication Date: 2019-08-23
SUN YAT SEN UNIV
View PDF3 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Second, traditional patent classification is done manually by experts in various fields of the patent office, which is time-consuming and labor-intensive. Research on automatic patent classification methods based on computer

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
  • Chinese patent classification method and system for TRIZ invention principle and storage medium
  • Chinese patent classification method and system for TRIZ invention principle and storage medium
  • Chinese patent classification method and system for TRIZ invention principle and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0037] Such as figure 1 As shown, the scheme provided by the present invention proposes a Chinese patent classification method for TRIZ invention principles, which includes the following steps:

[0038] S1. Obtain patent text data to construct a training data set, and mark the patent text in the training data set based on the TRIZ invention principle;

[0039] S2. Project the words of the patent text in the training data set to a low-dimensional vector space to obtain a word vector representation;

[0040] S3. Build a Bi-LSTM model, and input the word vector representation of the patent text in the training data set into the Bi-LSTM model to train it;

[0041] S4. Obtain the patent text to be classified, project the words of the patent text to be classified into a low-dimensional vector space, obtain the word vector representation, input the word vector representation into the Bi-LSTM model, and the Bi-LSTM model outputs the text to be classified Classification results of pa...

Embodiment 2

[0096] This embodiment provides a system based on the method of Embodiment 1, which includes a training data set construction module, a TRIZ invention principle labeling module, a word vector representation projection module, a Bi-LSTM model construction module, and a patent text acquisition module to be classified; wherein The training data set construction module is used to construct the training data set; the TRIZ invention principle labeling module is used to label the patent text in the training data set based on the TRIZ invention principle; the word vector representation projection module is used to combine the patent text in the training data set and the The words of the classified patent text are projected into the low-dimensional vector space to obtain the word vector representation; the Bi-LSTM model building block is used to construct the Bi-LSTM model, and the Bi-LSTM model is trained using the word vector representation of the patent text in the training data set. ...

Embodiment 3

[0098] This embodiment provides a storage medium with a program stored therein, and the execution steps of the method for classifying Chinese patents in Embodiment 1 are executed when the program is running.

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 relates to a Chinese patent classification method oriented to a TRIZ invention principle, which comprises the following steps: acquiring patent text data to construct a training data set, and marking patent texts in the training data set based on the TRIZ invention principle; performing preprocessing of removing meaningless symbols, removing non-Chinese words, segmenting words and removing stop words on the patent text in the training data set, and then projecting the obtained words into a low-dimensional vector space to obtain word vector representation; constructing a Bi-LSTM model, inputting the word vector representation of the patent text in the training data set into the Bi-LSTM model to be trained; obtaining patent texts to be classified, projecting words of the patenttexts to be classified to a low-dimensional vector space to obtain word vector representations, and inputting the word vector representations to Bi-LSTM model, and enabling the Bi-LSTM model to output a classification result of the patent text to be classified.

Description

technical field [0001] The present invention relates to the technical field of patent text classification, and more specifically, relates to a Chinese patent classification method, system and storage medium oriented to TRIZ invention principles. Background technique [0002] As one of the most important manifestations of technological innovation and scientific research, patents contain a large amount of technical information. They are an important indicator to measure a country's or enterprise's technological innovation capabilities and a direct manifestation of its core competitiveness. Research on existing patent texts can not only obtain the technological development dynamics and level of a country, region or enterprise, excavate its technological development path and development law, but also guide the direction of future technological innovation. At the same time, it can provide reliable support for enterprise decision-making, shorten the innovation cycle, and reduce R&...

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
IPC IPC(8): G06F16/35G06N3/04G06N3/08
CPCG06N3/049G06N3/08G06F16/35
Inventor 黄芸茵常会友
Owner SUN YAT SEN UNIV
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