Training method of aresponse model, response method, device, equipment and storage medium

A technology of model training and training methods, applied in the field of artificial intelligence, can solve the problems of operation and maintenance knowledge transfer, time-consuming and labor-intensive, etc., and achieve the effect of improving similarity, versatility, and generalization ability

Pending Publication Date: 2021-06-11
CHINA CONSTRUCTION BANK
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

To sum up, the level of automated operation and maintenance is an important cornerstone of AIOps, and AIOps will be based on automated operation and maintenance, combining AI and operation and maintenance well, but the current AIOps still has

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
  • Training method of aresponse model, response method, device, equipment and storage medium
  • Training method of aresponse model, response method, device, equipment and storage medium
  • Training method of aresponse model, response method, device, equipment and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0038] figure 1 It is a schematic flowchart of a method for training a response model provided in Embodiment 1 of the present invention. This embodiment is applicable to the situation of training the answering model, especially applicable to the situation of training the answering module in the operation and maintenance robot. The method can be executed by a response model training device, and the response model training device can be implemented in software and / or hardware, for example, the response model training device can be configured in a computer device. Such as figure 1 As shown, the method includes:

[0039] S110. Construct an intent training sample based on the entity information extracted from the knowledge graph.

[0040] In this embodiment, considering the migration of operation and maintenance knowledge, avoiding new scenarios and new data, the technical problems of intelligent operation and maintenance need to be rebuilt from scratch, automatically generating...

Embodiment 2

[0061] figure 2 It is a schematic flowchart of a method for training a response model provided in Embodiment 2 of the present invention. On the basis of the foregoing embodiments, this embodiment optimizes the training of the pre-built response model by using the response model training samples. Such as figure 2 As shown, the method includes:

[0062] S210. Construct an intent training sample based on the entity information extracted from the knowledge map.

[0063] S220. Construct a question answering training sample based on the pre-trained model.

[0064] S230. Construct a response model training sample according to the intent training sample and the question-and-answer training sample.

[0065] S240. Input the response model training samples into the feature extraction module, and obtain the initial features of the samples output by the feature extraction module.

[0066] In this embodiment, the response model includes a feature extraction module, an intention ident...

Embodiment 3

[0078] image 3 It is a schematic flowchart of a method for training a response model provided in Embodiment 3 of the present invention. On the basis of the foregoing embodiments, this embodiment optimizes the training of the pre-built response model by using the response model training samples. Such as image 3 As shown, the method includes:

[0079] S310. Construct an intent training sample based on the entity information extracted from the knowledge map.

[0080] S320. Construct a question answering training sample based on the pre-trained model.

[0081] S330. Construct a response model training sample according to the intent training sample and the question-and-answer training sample.

[0082] S340. Input the response model training samples into the feature extraction module, and obtain initial features of the samples output by the feature extraction module.

[0083] S350. Input the initial feature of the sample into the intent recognition module, and obtain the samp...

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 embodiment of the invention relates to the technical field of artificial intelligence, and discloses a training method of a response model, a response method, a device, equipment and a storage medium, and the training method of the response model comprises the steps: constructing an intention training sample based on entity information extracted from a knowledge graph; constructing a question and answer training sample based on the pre-training model; constructing a response model training sample according to the intention training sample and the question and answer training sample; and training a pre-constructed response model by using the response model training sample to obtain a trained response model. According to the method provided by the embodiment of the invention, the training corpus is automatically generated according to the knowledge graph extraction result, session design and session construction of the end-to-end response model are completed, and the similarity between the generated response model training sample and the real question and answer data is improved by combining the rules and deep learning; and the generalization ability of the generated text is improved to a certain extent, so that the universality of the response model is improved.

Description

technical field [0001] Embodiments of the present invention relate to the field of artificial intelligence, and in particular, relate to a training method of a response model, a response method, a device, a device, and a storage medium. Background technique [0002] AIOps (Artificial Intelligence for IT Operations, AIOps) combines the capabilities of artificial intelligence with operation and maintenance, and improves the efficiency of operation and maintenance through machine learning. AIops advocates that machine learning algorithms can automatically learn from massive operation and maintenance data (including the event itself and manual processing logs of operation and maintenance personnel), and continuously refine and summarize rules. On the basis of automated operation and maintenance, AIOps adds a brain based on machine learning. The command monitoring system collects the data required for brain decision-making, makes analysis and decision-making, and directs automate...

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): G06K9/62G06F16/36G06F40/35G06N3/08
CPCG06F16/367G06F40/35G06N3/08G06F18/214
Inventor 张美伟李昱王全礼张晨杨占栋
Owner CHINA CONSTRUCTION BANK
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