Unlock instant, AI-driven research and patent intelligence for your innovation.

Relationship extraction method and device based on semantic level

A relation extraction and semantic technology, applied in semantic analysis, natural language data processing, biological neural network model, etc., can solve the problems of inability to classify named entities by entity vocabulary, relation extraction error, inability to identify entity vocabulary, etc.

Active Publication Date: 2021-12-24
THE 28TH RES INST OF CHINA ELECTRONICS TECH GROUP CORP
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Most of the technical solutions in the prior art are carried out based on English corpus, but Chinese and English have great differences in linguistics, Chinese expressions are more diverse, and there are a large number of different words expressing the same meaning, so Traditional relational extraction schemes do not work well in Chinese. They cannot accurately identify entity words from the text, and cannot classify entity words into correct named entities. Therefore, it is difficult to obtain effective information from the text, and there are problems in relation extraction. error problem, while relational triples are far from being able to fully express the overall meaning of the Chinese text

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
  • Relationship extraction method and device based on semantic level
  • Relationship extraction method and device based on semantic level

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0021] The technical solution of the present invention will be further described below in conjunction with the accompanying drawings.

[0022] refer to figure 1 , which is a schematic flowchart of the semantic level-based relation extraction method provided by the present invention.

[0023] Step 1. Establish a named entity dictionary associated with the domain of the dataset.

[0024] In the embodiment of the present invention, the named entity dictionary includes named entities and corresponding entity vocabulary, semantic levels of named entities, and named entity matching relationships in the semantic levels; the data set includes a training set.

[0025] In the specific implementation, since there are a large number of different words that express the same meaning in Chinese, the field of the named entity dictionary established needs to be associated with the field of the text to be recognized, and all the words in this field can be covered as much as possible. Internal...

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 discloses a relationship extraction method and device based on a semantic level. The method comprises the following steps: establishing a named entity dictionary associated with a data set field; and performing entity vocabulary relation extraction on the to-be-recognized text by using the trained relation extraction model to obtain relation statements of entity vocabularies. The training process of the relation extraction model comprises: training the relation extraction model to recognize the relation between the entity vocabularies and the named entities in the training set text, and performing entity vocabulary relationship training on the relationship extraction model by using the training set based on the semantic level. By adopting the technical scheme, the entity vocabularies in the text data are effectively recognized, and the extraction result can completely and accurately express the overall meaning of the Chinese text.

Description

technical field [0001] The present invention relates to the technical field of natural language processing, in particular to a semantic level-based relation extraction method and device. Background technique [0002] Nowadays, the amount of data on the Internet is growing explosively, but most of them are unstructured data, and it is difficult for users to quickly obtain effective information from them. Therefore, converting unstructured data into structured data and realizing the rapid extraction of effective information has become one of the important research topics now. [0003] The technical solutions commonly used in the existing technology involve named entity recognition and relationship extraction. Among them, named entity recognition (Named Entity Recognition, NER) is a key technology in the process of knowledge map construction, which mainly completes the recognition of entity vocabulary from unstructured data. , and classified into predefined named entity types,...

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): G06F40/30G06F40/295G06F40/242G06F16/33G06K9/62G06N3/04
CPCG06F40/30G06F40/242G06F40/295G06F16/334G06N3/044G06F18/214
Inventor 方酉后弘毅郭嘉欣
Owner THE 28TH RES INST OF CHINA ELECTRONICS TECH GROUP CORP