Defect automatic rating and disposal suggestion pushing method based on natural language processing

A natural language processing and defect technology, applied in data processing applications, electrical digital data processing, special data processing applications, etc., can solve problems such as unclear expression, low efficiency, and impact on classification accuracy, so as to improve efficiency and accuracy , a large amount of reference data, and the effect of improving accuracy

Pending Publication Date: 2022-04-29
STATE GRID FUJIAN ELECTRIC POWER CO LTD +1
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] However, the manually recorded defect text itself may have quality problems such as unclear expression, which will affect the mining effect
In addition, the current defect severity classification relies on manual completion, which is not only inefficient, but also when encountering complex problems, it will be difficult to make judgments due to knowledge structure and experience limitations, and the classification accuracy will be affected

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
  • Defect automatic rating and disposal suggestion pushing method based on natural language processing
  • Defect automatic rating and disposal suggestion pushing method based on natural language processing
  • Defect automatic rating and disposal suggestion pushing method based on natural language processing

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0046] The automatic defect rating and disposal suggestion push method based on natural language processing includes the following steps:

[0047] Obtain the text to be rated, the lower limit of rating similarity, and the lower limit of similarity of disposal suggestions; the text to be rated is a description of the defect;

[0048] The text to be rated is matched by a tree path matching algorithm to obtain a matching classification standard A, and a rating result is output according to the classification standard A;

[0049] The text to be rated is matched with the historical defect library through a text similarity matching algorithm to obtain the classification standard B that meets the lower limit of the rating similarity, and the rating result is output according to the classification standard B; the historical defect library includes various The historical description text of the defect; the historical description text is the description text of various defects by variou...

Embodiment 2

[0053] refer to Figure 1-2 , a defect automatic rating and disposal suggestion push method based on natural language processing. On the basis of Embodiment 1, the text to be rated is matched through a tree path matching algorithm to obtain a matching classification standard A. According to the classification standard A outputs the rating results, specifically:

[0054] S1. Establish a standard tree structure, the standard tree structure includes several layers, the first layer is a root node, and each upper layer node corresponds to several sub-nodes located in the lower layer;

[0055] S2. Taking the root node as a candidate node;

[0056] S3. Perform similarity matching between each child node of the candidate node and the text to be rated;

[0057] S4. Judging whether there is a child node A matching the text to be rated; if it exists, each of the child nodes A is used as a candidate node; if it does not exist, each child node of the candidate node is As an alternative no...

Embodiment 3

[0069] Based on the natural language processing-based defect automatic rating and disposal suggestion push method, on the basis of the first embodiment, the text to be rated is matched with the historical defect library through a text similarity matching algorithm, and the result that satisfies the lower limit of the rating similarity is obtained Classification criteria B, specifically:

[0070] Obtain the distance sen_dis1 between the text sen1 to be rated and each historical description text sen2 in the historical defect library:

[0071]

[0072] Wherein, diff(sen1, sen2) is the text length that the text sen1 to be rated does not exist in the historical description text sen2, and len1 is the length of the text sen1 to be rated;

[0073] Screening out the historical description text sen2 corresponding to the distance sen_dis1 that satisfies the lower limit of the rating similarity; obtaining the classification criteria corresponding to each of the historical description t...

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 relates to an automatic defect rating and disposal suggestion pushing method based on natural language processing. The method comprises the following steps: acquiring a defective text to be rated, a rating similarity lower limit and a disposal suggestion similarity lower limit; matching the to-be-rated text through a tree path matching algorithm to obtain a matched classification standard A, and outputting a rating result according to the classification standard A; matching the to-be-rated text with the historical defect library through a text similarity matching algorithm to obtain a classification standard B meeting the lower limit of rating similarity, and outputting a rating result according to the classification standard B; and matching the to-be-rated text with the disposal suggestion library through a text similarity matching algorithm to obtain disposal suggestions meeting the lower limit of the similarity of the disposal suggestions, and outputting the disposal suggestions. According to the method, automatic classification of the defective text is achieved, and due to the fact that the method is based on the defect standard and the historical defect data, the rating basis is reliable, the reference data amount is large, and the classification efficiency and accuracy of the defective text can be effectively improved.

Description

technical field [0001] The invention relates to a defect automatic rating and disposal suggestion push method based on natural language processing, and belongs to the technical field of electric equipment auxiliary maintenance. Background technique [0002] With the continuous construction and development of smart grid and ubiquitous power Internet of Things, new technologies and tools are gradually introduced in the operation and maintenance process of the power system, and it is necessary to improve or optimize the original work process and improve work efficiency. Take the defects found during equipment inspection as an example. At present, inspectors manually record them in detail, judge the severity of the defects, and then input them into the corresponding system centrally. The equipment defects recorded in detail by these staff during the operation and maintenance process are called power equipment defect texts. The defect text usually includes three parts: the defec...

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): G06F16/35G06F16/335G06K9/62G06Q50/06
CPCG06F16/35G06F16/335G06Q50/06G06F18/22G06F18/24
Inventor 陈丽霞上官诚江郑鹭洲李怀蔡继东王昕张登灵郑翔何剑锋
Owner STATE GRID FUJIAN ELECTRIC POWER 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