BERT-based government affair official document ontology concept extraction method

A technology of official documents and ontology, which is applied in the field of ontology learning, can solve problems such as insufficient comprehensiveness and richness of ontology concepts, poor extraction of ontology concepts, etc., and achieve the effect of improving the clustering effect

Pending Publication Date: 2019-12-13
CETC BIGDATA RES INST CO LTD
View PDF7 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] In order to solve the above technical problems, the present invention provides a BERT-based method for extracting concepts of government official texts. The BERT-based method for extracting concepts of government official texts aims at the lack of comprehensive and rich ontology concepts in the process of building ontology knowledge bases in the government domain and the existing In order to solve the problem of ineffective extraction of ontology concepts, through the integration of natural language processing, linguistics, statistics, deep learning, and cutting-edge BERT word vector representation and other technologies, the accuracy of concept extraction of government official texts is improved, and the concept of government official texts is further enriched content

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
  • BERT-based government affair official document ontology concept extraction method
  • BERT-based government affair official document ontology concept extraction method
  • BERT-based government affair official document ontology concept extraction method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0056] As mentioned above, a BERT-based method for extracting the concept of government official texts includes the following steps:

[0057] (1) Using the crawler method to obtain government official document data from the website of the people's government of a certain province using crawlers;

[0058] (2) Perform text data preprocessing on the public government official document data;

[0059](3) Terminological rules extracted according to linguistic regulations;

[0060] (4) Use the C-value algorithm to realize the term extraction of government official document data;

[0061] (5) Use BERT to realize the vectorized representation of official document terms;

[0062] (6) Use the contour coefficient method to estimate the number of official document term classes;

[0063] (7) Use the K-means++ algorithm to realize the clustering of official document terms;

[0064] (8) Select Top_N in the class as the official document concept to realize the extraction of official docume...

Embodiment 2

[0067] Such as figure 1 As shown, a BERT-based method for extracting the concept of government official texts includes the following steps:

[0068] First execute step S1 to obtain government official document data

[0069] Send a request to the opposite server through the URL, obtain the static or dynamic code of the page, and clean the required document title and document text from the page code by parsing the DOM tree or other aspects to realize data capture and storage.

[0070] This example crawls the official document data of a certain province's government affairs disclosure from 2013 to 2018, with a total of 8,530 official document documents and 9,796,600 official document words.

[0071] Next, execute step S2 to perform text preprocessing on all official government documents

[0072] Use the python development language to remove stop words, numbers, English characters, punctuation marks, low-frequency words, etc., and then use the stutter word segmentation tool to c...

Embodiment 3

[0106] As mentioned above, a BERT-based method for extracting the concept of government official texts includes the following steps:

[0107] (1) Obtain government official document data;

[0108] (2) Perform text data preprocessing on the public government official document data;

[0109] (3) Restrict the terminology of official documents with linguistic rules;

[0110] (4) Extract terminology from government official document data;

[0111] (5) Carry out vectorized representation to the extracted official document terms;

[0112] (6) Estimate the number of official document term categories;

[0113] (7) Realize official document term clustering by combining the clustering method with the vector representation of terms;

[0114] (8) Select official document concepts from each type of term set to complete the extraction of official document body concepts;

[0115] (9) Evaluate and verify the concept extraction results.

[0116] In the step (1), the government official do...

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 BERT-based government affair official document ontology concept extraction method. The method comprises the following steps of (1) obtaining government affair official document data; (2) performing text data preprocessing on the public government affair official document data; (3) establishing linguistic rules of terms; (4) performing official document ontology term extraction; (5) estimating the category number of the official document ontology terms; (6) carrying out word vector representation for the official document ontology terms; (7) completing term clustering;(8) extracting an official document ontology concept; and (9) realizing evaluation and verification of an ontology concept extraction effect. Effective technical means of government affair work are overall planned, powerful support and guarantee are provided for application of government affair official affairs such as sharing exchange, information retrieval, information extraction and governmentaffair atlas construction, the clustering effect of official document terms is improved, and solid guarantee and support are provided for precision of official document ontology concept extraction.

Description

technical field [0001] The invention relates to a BERT-based method for extracting concepts of government official texts, and belongs to the technical field of ontology learning. Background technique [0002] Ontology is a normative and clear definition of the concept form and the relationship between concepts. With the rapid development of the Internet and information technology, ontology, as an important conceptual layer semantic resource knowledge base, provides strong support for the wide application of knowledge graph, question answering system, information retrieval, information extraction and other technologies, and ontology learning has become a A hot research topic in academia and engineering. [0003] Ontology learning refers to the construction process of ontology knowledge base. Ontology learning mainly includes the following four stages: (1) term extraction; (2) concept extraction; (3) relation extraction; (4) ontology formation. Generally speaking, ontology c...

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): G06F16/9032G06F16/9035G06F16/906G06F16/951G06F17/27G06Q50/26
CPCG06F16/90332G06F16/9035G06F16/906G06F16/951G06Q50/26
Inventor 闫盈盈王进曹扬阚丹会丁剑飞张婧慧
Owner CETC BIGDATA RES INST 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