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

Multi-tag entity-relationship joint extraction method based on depth neural network and annotation strategy

A deep neural network and joint extraction technology, which is applied in the field of multi-label entity-relationship joint extraction based on deep neural network and labeling strategy, can solve problems such as failure to solve relationship overlap well, failure to achieve joint extraction, etc.

Active Publication Date: 2019-03-29
XI AN JIAOTONG UNIV
View PDF2 Cites 45 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, most of the existing joint extraction methods are based on features or shared parameters, which do not achieve joint extraction in the true sense, and do not solve the problem of overlapping relations well, that is, an entity may have multiple relation labels

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
  • Multi-tag entity-relationship joint extraction method based on depth neural network and annotation strategy
  • Multi-tag entity-relationship joint extraction method based on depth neural network and annotation strategy
  • Multi-tag entity-relationship joint extraction method based on depth neural network and annotation strategy

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0044] Embodiment: It is mainly used to extract the entities in each sentence of the data set and the semantic relationship of the entities. Both training data and test data are selected from the NYT dataset.

[0045] Such as figure 1 The process shown and figure 2 As shown in the model, the method of the present invention comprises the following steps,

[0046] Step 1: Firstly, word segmentation is performed on the training text and test text, and the training text obtained after word segmentation is marked with a marking strategy. The marking strategy is specifically: according to the labeling of the training text, set an "O" label (not belonging to any relationship) or a "non-O" label (having a relationship) for each word according to the labeling of the training text. The non-O label consists of three parts: word position, relation category and relation role. Wherein, the word position marks include B (begin), I (inside), E (end) and S (single), which are used to repr...

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 multi-tag entity based on a depth neural network and a tagging strategy. Relation joint extraction method based on annotation strategy can effectively avoid the error in the named entity recognition subtask will propagate to the relational classification subtask and ignore the interaction between the two subtasks. At that same time, the invention useS Tree-GRU, as the coding layer, can make the model learn the syntactic information of the whole sentence more fully, and help to identify the relationship between the two entities and the type of the relationship. In addition, because of the complexity of the massive text itself, an entity may be included in a variety of relationships, the use of multi-tag classifiers, a good solution to the above problem. The method of the invention obtains good results in different data fields, and can efficiently, accurately and intelligently extract information with practical value and research significance from massive text data.

Description

technical field [0001] The invention relates to the field of natural language processing, in particular to a multi-label entity-relationship joint extraction method based on a deep neural network and a labeling strategy. Background technique [0002] In the era of big data and artificial intelligence, information extraction and semantic understanding have become the focus of researchers in recent years. Extracting the semantic relationship between entities and entity pairs in unstructured text is one of the important tasks of information extraction, and it is also the key to achieve semantic understanding. However, unstructured text data expressed in natural language has the characteristics of huge data volume, complex structure, and fast generation speed. It is very difficult for relevant researchers to quickly and accurately obtain valuable knowledge and information from a large amount of text. Therefore, how to extract valuable information from massive data intelligently...

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): G06F17/27G06N3/04
CPCG06N3/049G06F40/211G06F40/295
Inventor 李辰龙雨王轩
Owner XI AN JIAOTONG UNIV
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