Multi-strategy fused method and device for identifying named entity

A named entity recognition and named entity technology, applied in instrumentation, computing, semantic analysis, etc., can solve the problems of inability to assign named entity roles, low named entity recognition accuracy and recall rate, etc.

Active Publication Date: 2017-11-07
ZHONGKE DINGFU BEIJING TECH DEV
View PDF11 Cites 44 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0009] This application provides a multi-strategy fusion named entity recognition method and device to solve the problem of low accuracy and recall of na...

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-strategy fused method and device for identifying named entity
  • Multi-strategy fused method and device for identifying named entity
  • Multi-strategy fused method and device for identifying named entity

Examples

Experimental program
Comparison scheme
Effect test

experiment example 1

[0177] Influence of mode value on F value in the second recognition of Experimental Example 1

[0178] In the second recognition step used in this experimental example, the preset values ​​are different, and the final named entity recognition results are significantly different. This experimental example investigates the influence of the preset values ​​on the named entity recognition results.

[0179] The preset value is the mode of the at least two recognition models;

[0180] The named entity recognition result is measured by the F value, and the higher the F value, the more reliable the recognition result is, wherein,

[0181] Accuracy (P) = number of named entities recognized correctly / number of named entities recognized by the machine,

[0182] Recall rate (R) = number of correctly identified named entities / number of named entities in the test corpus.

[0183] F value = 2*P*R / (P+R).

[0184] The recognition models used in the second recognition in this experimental ...

experiment example 2

[0193] Experimental example 2 Named entity recognition results when each recognition model is used alone

[0194] This experiment example tests the results of using one recognition model alone for named entity recognition to compare the reliability of the two named entity recognition methods of single recognition model and multi-recognition model fusion.

[0195] The recognition models used in this experimental example are the CRF recognition model used in the primary recognition, the precise word segmentation algorithm used in the second recognition, the word segmentation algorithm with new word discovery function, and the named entity recognition algorithm of structured perceptron. The result is as Figure 6 and shown in Table 2.

[0196] Table 2 Reliability of single recognition model named entity recognition method

[0197]

[0198] exist Figure 6 Among them, line A is the line of recall rate corresponding to each recognition method; line B shows the line of F val...

experiment example 3

[0200] Experimental example 3 Named entity recognition results of each recognition model of the application method

[0201] In this experimental example, the accuracy rate, recall rate and F value of the first recognition result, the second recognition result and the third recognition result were calculated respectively by using the method provided in this application. The results are shown in Table 3 below.

[0202] Table 3 Named entity recognition results of each recognition model of the application method

[0203]

[0204] It can be seen from Table 3 that according to the method provided by the present application, the accuracy, recall and F value of the third recognition result obtained on the basis of the first recognition result and the second recognition result have been greatly improved, that is, , the method provided by this application can cope with new situations such as massive data scale, diversified entity types, and endless new words, and has a high recall ...

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 multi-strategy fused method and device for identifying a named entity. A first identification result is obtained by utilizing a first identifying model to identify the named entity in obtained linguistic data, the first identifying model can update and extend a corpus in the method, accordingly the newly produced named entity in the linguistic data can be identified, thus a first identifying result has higher accuracy rate, the named entity in the linguistic data is identified by utilizing a multi-identifying-model fused method to obtain a second identifying result, the first identifying result and the second identifying result are fused to obtain a third identifying result, then a semantic mining system is utilized to conduct role assignment on the named entity, the named entity having a role is output, accordingly it is achieved that the named entity is reliably identified under the situations that data is massive, the entity type is diversified and new words emerge in endlessly, and role assignment is conducted on the identified named entity.

Description

technical field [0001] The present application relates to the field of natural language processing, in particular to a method and device for identifying named entities with multi-strategy fusion. Background technique [0002] Named entities are people's names, organization names, place names, and all other entities identified by names. They are the basic information elements in the text, an important carrier of information expression, and the basis for correct understanding and processing of text information. Chinese named entity recognition is one of the basic tasks in the field of natural language processing. Its main task is to identify and classify the name entities and meaningful quantitative phrases that appear in the text, mainly including names of people, places, organizations, and time. expressions, dates, numeric expressions, etc. [0003] In terms of natural language processing research, named entity recognition plays an important role in the application fields o...

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): G06F17/30G06F17/27
CPCG06F16/353G06F16/367G06F40/295G06F40/30
Inventor 赵红红王萌萌晋耀红蒋宏飞杨凯程
Owner ZHONGKE DINGFU BEIJING TECH DEV
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