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

Named entity recognition method, device and system

A named entity recognition and named entity technology, applied in special data processing applications, instruments, electrical digital data processing, etc. The effect of improving versatility and improving adaptability

Inactive Publication Date: 2018-07-06
ULTRAPOWER SOFTWARE
View PDF4 Cites 21 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] However, in the prior art entity naming recognition method based on the CRF model, the process of training, management and use of the model is carried out within a specific tool or framework, therefore, the prior art entity naming recognition method based on the CRF model is The adaptability to different corpora during the training process is poor, and it cannot be universally applied to various named entity recognition application scenarios

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
  • Named entity recognition method, device and system
  • Named entity recognition method, device and system
  • Named entity recognition method, device and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0045] The embodiment of the present application provides a named entity recognition method, see figure 1 , which is a flow chart of a named entity recognition method provided in the embodiment of the present application, the method includes the following steps:

[0046] Step S110, labeling the training corpus according to the preset labeling specification; the preset labeling specification presets the type and labeling method of the named entity.

[0047] In this application, according to the user's specific needs for named entity recognition in different business scenarios, the training corpus with different corpus characteristics is selected. For example, when the user's business scenario requires named entity recognition for political news texts, the The selected training corpus should have the content characteristics of current affairs news texts.

[0048] In this application, in order to facilitate the management of training corpus applicable to different business scena...

Embodiment 2

[0217] An embodiment of the present application provides a named entity recognition device, Figure 10 A structural block diagram of a named entity recognition device provided in the embodiment of the present application, such as Figure 10 As shown, the device includes:

[0218] The labeling unit 210 is used to label the training corpus according to the preset labeling specification; the preset labeling specification defines the type and labeling method of the named entity;

[0219] The first conversion unit 221 is configured to perform a first format conversion on the labeled training corpus to generate a first standard corpus;

[0220] A first training unit 231, configured to train a first CRF model according to the first standard corpus;

[0221] The first evaluation unit 241 is configured to evaluate the predictive ability of the first CRF model after training, and add the first CRF model that meets the predictive ability requirements to the list of available models;

...

Embodiment 3

[0228] The embodiment of the present application provides a named entity recognition system, Figure 11 A schematic structural diagram of a named entity recognition system provided by the embodiment of this application, such as Figure 11 As shown, the system includes:

[0229] memory 310 and processor 320;

[0230] The memory 310 is used to store the executable program of the processor 320;

[0231] The processor 320 is configured to perform the following program steps:

[0232] Annotate the training corpus according to the preset label specification; the preset label specification defines the type and labeling method of the named entity;

[0233] Perform format conversion on the labeled training corpus, the format conversion includes first format conversion, and / or, second format conversion; the first format conversion generates the first standard corpus, and the second format conversion generating a second standard corpus;

[0234] training a first CRF model according ...

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

An embodiment of the application provides a named entity recognition method, device and system. Two CRF (conditional random field) models are integratedly provided, so that a user may select differentCRF models according to actual scene needs and application environments to perform model training and named entity predicting; in addition, two methods to train CRF models are provided for the firstand second CRF models, and the need for different training on CRF models under different training corpus scales is met. In addition, a unified preset marking guideline is provided for the two CRF model training methods; the type and marking modes of named entities are uniformly defined through the preset marking guideline; with any CRF model for training or predicting, a user may provide uniform marking for training corpuses according to the unified preset marking guideline, and setting a special marking mode for each CRF model is not required. Therefore, the ability of the CRF models to adaptto different training corpuses is improved, and CRF model universality is improved.

Description

technical field [0001] The present application relates to the technical field of natural language processing, in particular to a named entity recognition method, device and system. Background technique [0002] Named Entity Recognition (NER for short), also known as "proper name recognition", refers to the recognition of entities with specific meanings in text, mainly including names of people, places, institutions, and proper nouns. In the field of natural language processing, named entity recognition is an important basic tool in application fields such as information extraction, question answering system, syntactic analysis, machine translation, and semantic web-oriented metadata annotation. It plays an important role in the process of natural language processing technology becoming practical. status. [0003] Named entity recognition methods in the prior art mainly include: rule-based and dictionary-based named entity recognition methods, statistics-based entity naming ...

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): G06F17/27
CPCG06F40/295
Inventor 李德彦晋耀红席丽娜
Owner ULTRAPOWER SOFTWARE
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