License entity extraction method and system based on deep learning

An entity extraction, deep learning technology, applied in neural learning methods, instruments, unstructured text data retrieval, etc.

Pending Publication Date: 2021-05-28
SHANDONG LANGCHAO YUNTOU INFORMATION TECH CO LTD
View PDF1 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The technical task of the present invention is to provide a method and system for extracting license entities based on deep learning to solve how to improve the accuracy of license nouns and work behaviors in intelligent question-and-answer services, so as to ensure more accurate replies to questions, thereby improving service quality and user satisfaction

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
  • License entity extraction method and system based on deep learning
  • License entity extraction method and system based on deep learning
  • License entity extraction method and system based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0074] as attached figure 1 As shown, the deep learning-based license entity extraction method of the present invention is to identify the two entities, the license name and the service behavior in the user's consultation questions, through the BiLSTM-CRF model in the intelligent robot question-and-answer service in the government affairs system and extraction to improve the accuracy of certificate names and work behaviors in the intelligent question-and-answer service; the details are as follows:

[0075] S1, generating a training data set;

[0076] S2. Training the BiLSTM-CRF model for extracting license nouns and service behavior entities;

[0077] S3. Test the training effect of the BiLSTM-CRF model.

[0078] In this embodiment, the training data set of step S1 includes two entity categories, namely license and behavior; five labels are included in the training data set, which are respectively B-licence, I-licence, B-behavior, I-behavior and labels O; In an example such...

Embodiment 2

[0103] The license entity extraction system based on deep learning of the present invention, the system includes,

[0104] The generation module is used to generate the training data set, and use the script to automatically mark and generate the training data based on the BIO data mark format;

[0105] The training module is used to train the BiLSTM-CRF model for extracting certificate nouns and service behavior entities; among them, the BiLSTM-CRF model is divided into three layers, as follows:

[0106] The first layer is the Embedding layer: use the pre-trained or randomly initialized embedding matrix to map each word in the sentence from a one-hot vector to a low-dimensional dense word vector;

[0107] The second layer is a two-way LSTM layer: automatic extraction of sentence features;

[0108] The third layer is the CRF layer: for sequence labeling of sentences;

[0109] The test module is used to test the training effect of the BiLSTM-CRF model.

[0110] The training d...

Embodiment 3

[0129] An embodiment of the present invention also provides a computer-readable storage medium, in which a plurality of instructions are stored, and the instructions are loaded by a processor, so that the processor executes the deep learning-based certificate entity extraction method in any embodiment of the present invention. Specifically, a system or device equipped with a storage medium may be provided, on which a software program code for realizing the functions of any of the above embodiments is stored, and the computer (or CPU or MPU of the system or device) ) to read and execute the program code stored in the storage medium.

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 license entity extraction method and system based on deep learning, belongs to the field of natural language processing, and aims to solve the technical problem of how to improve the accuracy of license nouns and handling behaviors in an intelligent question and answer service and ensure more accurate reply to questions so as to improve the service quality and the satisfaction degree of a user. According to the technical scheme, the method specifically comprises the following steps: generating a training data set; training a BiLSTM-CRF model for extracting the license nouns and the affair handling behavior entities; and testing the training effect of the BiLSTM-CRF model. The system comprises a generation module, a training module and a test module. The training module is used for training a BiLSTM-CRF model extracted by a license noun and an affair handling behavior entity; the BiLSTM-CRF model is divided into three layers, the first layer is an Embedding layer, and each word in a sentence is mapped into a low-dimensional dense word vector from a one-hot vector by using a pre-trained or randomly initialized embedding matrix; the second layer is a bidirectional LSTM layer for automatically extracting sentence features; and the third layer is a CRF layer for performing sequence labeling of statements.

Description

technical field [0001] The invention belongs to the field of natural language processing, and relates to technologies such as sequence tagging, entity recognition, word segmentation and word frequency statistics, and specifically a method and system for extracting certificate entities based on deep learning. Background technique [0002] With the continuous deepening of "Internet + government services", government intelligent question-answering robots have been widely used in large government agencies, legal centers, tax centers, bank halls and other public places, and various functions, comments and knowledge can be customized according to different industries and occasions , the government intelligent question answering robot can effectively help the masses solve various government affairs, help staff solve many repetitive problems, improve work efficiency and service level, and is deeply loved by major institutions. [0003] In intelligent question-and-answer services, qu...

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): G06F40/295G06F40/211G06F16/332G06F16/35G06N3/04G06N3/08
CPCG06F40/295G06F40/211G06F16/353G06F16/3329G06N3/08G06N3/044G06N3/045
Inventor 马凤强
Owner SHANDONG LANGCHAO YUNTOU INFORMATION TECH CO LTD
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