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

Entity recognition model training method and device, entity recognition method and device and terminal equipment

A technology of entity recognition and training method, applied in the computer field to achieve the effect of improving accuracy and avoiding recognition

Active Publication Date: 2020-11-10
CCB FINTECH CO LTD
View PDF4 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] At present, when identifying entities in text information, there is a lack of identification of meaningful new entities in various fields

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
  • Entity recognition model training method and device, entity recognition method and device and terminal equipment
  • Entity recognition model training method and device, entity recognition method and device and terminal equipment
  • Entity recognition model training method and device, entity recognition method and device and terminal equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0048] figure 1 It is a schematic flowchart of a method for training an entity recognition model provided in Embodiment 1 of the present invention. The method is applicable to the training of an entity recognition model. The method can be executed by a training device for an entity recognition model, wherein the device can be It is realized by software and / or hardware, and is generally integrated on a terminal device. In this embodiment, the terminal device includes but is not limited to: computers, mobile phones and other devices.

[0049] The present invention proposes a new entity discovery method for supervised learning parameter correction based on the idea of ​​seed iteration on a small data set, that is, a sub-information set, which can obtain a better-performing entity recognition model with a relatively small labor cost , and then discover more new entities in the text corpus.

[0050] Such as figure 1 As shown, a method for training an entity recognition model prov...

Embodiment 2

[0097] figure 2 It is a schematic flowchart of an entity recognition method provided in Embodiment 2 of the present invention, which is applicable to the situation of entity recognition, and the method can be executed by an entity recognition device, wherein the device can be implemented by software and / or hardware , and generally integrated on the terminal device. In this embodiment, the terminal device includes but not limited to computers, mobile phones and other devices.

[0098] Such as figure 2 As shown, an entity recognition method provided by Embodiment 2 of the present invention includes the following steps:

[0099] S210. Obtain text information.

[0100] Text information can be considered as information that characterizes the text to be predicted. The method for obtaining text information in this step is not limited, for example, the text information input by the user may be obtained, so as to input the text information into the trained entity recognition model...

Embodiment 3

[0149] image 3 A schematic structural diagram of a training device for an entity recognition model provided in Embodiment 3 of the present invention, which is applicable to the training of an entity recognition model, wherein the device can be implemented by software and / or hardware, and is generally integrated in a terminal device superior.

[0150] Such as image 3 As shown, the device includes:

[0151] An acquisition module 31, configured to acquire a text information set and a pre-built seed dictionary, the seed dictionary includes entities in the text information included in the text information set;

[0152] A splitting module 32, configured to split the text information set into at least two sub-information sets;

[0153] The iteration module 33 is used to perform iterations of entity recognition model training and prediction according to the seed dictionary and each sub-information set, and the recognition result after each round of iteration is used to update the...

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 an entity recognition model training method and device, an entity recognition method and device and terminal equipment. The method comprises the steps of obtaining a text information set and a pre-constructed seed dictionary, wherein the seed dictionary comprises entities in text information included in the text information set; splitting the text information set into at least two sub-information sets; according to the seed dictionary and each sub-information set, iterating entity recognition model training and prediction, wherein a recognition result after each round of iteration is used for updating the seed dictionary; and determining the entity recognition model after iteration as a trained entity recognition model. By means of the method, the situation that a new entity cannot be recognized is avoided, and the accuracy of entity recognition is improved.

Description

technical field [0001] The embodiments of the present invention relate to the field of computer technology, and in particular to training of entity recognition models, entity recognition methods, devices and terminal equipment. Background technique [0002] Entity refers to the word information with specific meaning in the text. At present, the entities involved in academics generally include 3 categories and 7 subcategories, of which 3 categories include entities, time and numbers, and 7 subcategories include names, Place name, organization name, time, date, currency and percentage. [0003] In addition to the 3 major categories and 7 subcategories of entity types defined in the academic field, there will be a category of entities in each technical field, which is a meaningful new entity in this field. For example, there are various financial entities in the financial field, such as names of financial institutions and names of financial products. [0004] At present, when...

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): G06F40/295G06F40/242
CPCG06F40/295G06F40/242
Inventor 袁晟君李宸庞帅付博
Owner CCB FINTECH CO LTD