Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Semantic relationship classification method capable of combining with multi-syntax structure

A technology of semantic relationship and classification method, which is applied in the field of semantic relationship classification that integrates multi-syntax structures to achieve the effect of improving robustness and classification efficiency

Active Publication Date: 2018-11-23
GUANGDONG UNIV OF TECH
View PDF10 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

At present, there is no robust method to combine the advantages of these three, so as to achieve a more stable and efficient classification result

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
  • Semantic relationship classification method capable of combining with multi-syntax structure
  • Semantic relationship classification method capable of combining with multi-syntax structure
  • Semantic relationship classification method capable of combining with multi-syntax structure

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0017] Such as Figure 1-5 As shown, the embodiment of the present invention discloses a kind of semantic relation classification method of fusion multi-syntax structure, comprises the following steps:

[0018] S1. Perform data preprocessing on the text sequence to obtain the text sequence S={S 1 ,S 2 ,...,S n}’s part-of-speech tag P={P 1 ,P 2 ,...,P n} and dependency label D = {D 1 ,D 2 ,...,D n}; In step S1, for the text sequence S={S to be input into the network 1 ,S 2 ,...,S n} to preprocess, and obtain the part-of-speech tag P={P of the text sequence by utilizing the Stanford parsing tool 1 ,P 2 ,...,P n} and dependency label D = {D 1 ,D 2 ,...,D n}, text sequence S={S 1 ,S 2 ,...,S n} in each character S n corresponds to a part-of-speech tag P n and relation label D n , are used as data input for network training.

[0019] In this step, the Stanford parsing tool is an existing text sequence parsing tool, and its purpose is to convert the text seque...

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 the technical field of natural language processing, in particular to a semantic relationship classification method capable of combining with a multi-syntax structure. The method comprises the following steps that: firstly, carrying out data preprocessing on a text sequence; then, carrying out bidirectional coding on the vector of the text sequence, utilizing an attention mechanism to carry out weighted learning on coding information, carrying out bidirectional coding on the weighted coding information through a CRF (Conditional Random Field), and obtaining entity tag information on the text sequence; and then, constructing a joint vector, constructing a complex network which combines with multiple syntaxes, utilizing shared parameters to carry out end-to-end training, obtaining an implicit state on a triple, carrying out splitting, and carrying out linear transformation to output a semantic relationship category. By use of the method, on the basis of a laminated recurrent neural network model, various syntax structures are combined, the problem that a single syntax structure model can not effectively adapt to other syntax structures is solved, so that different syntax structures can be effectively processed, the robustness of the model is improved, and classification efficiency is improved.

Description

technical field [0001] The present invention relates to the technical field of natural language processing, more specifically, relates to a kind of semantic relation classification method of fusion multi-syntax structure. Background technique [0002] With the development of the Internet, unstructured texts are growing exponentially. By using automated extraction tools, unstructured texts are transformed into structured knowledge, and these structured knowledge are applied to the construction of retrieval systems and knowledge bases. It can effectively help people improve work efficiency. Semantic relationship classification is one of the important technical means. [0003] Semantic relationship classification includes two subtasks of entity recognition and relationship classification, and the corresponding methods are based on sequence annotation and syntactic structure. The existing semantic relationship classification process mainly adopts a single specific syntactic st...

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): G06F17/30G06F17/27G06K9/62
CPCG06F40/211G06F40/30G06F18/25
Inventor 郝志峰陈培辉蔡瑞初温雯王丽娟陈炳丰
Owner GUANGDONG UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com