Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Machine self-learning construction knowledge atlas training method based on neural network

A neural network and knowledge map technology, applied in the field of neural network-based machine self-learning to build knowledge map training, can solve problems such as semantic understanding errors, limited robot knowledge base, unimaginable developers, etc., achieve small mean square error, improve The effect of the signal-to-noise ratio

Active Publication Date: 2017-06-20
吉林省盛创科技有限公司
View PDF6 Cites 14 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] However, in the current robot chat scenario, when the same or similar questions that match the user's requested question cannot be found in the robot knowledge base, the robot cannot return the correct or appropriate answer to the user.
[0004] In addition to the limited knowledge base of the robot, the limitation of the existing technology will also lead to errors in semantic understanding, thus making the user experience in the communication process with the robot poor.
In addition, the knowledge reasoning process of existing technologies also has certain limitations in knowledge reasoning. Traditional knowledge reasoning relies on program developers to write some rules to solve knowledge reasoning problems.
However, it is unimaginable for developers to exhaustively enumerate and formulate these rules

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
  • Machine self-learning construction knowledge atlas training method based on neural network
  • Machine self-learning construction knowledge atlas training method based on neural network
  • Machine self-learning construction knowledge atlas training method based on neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0047] The present invention will be further described in detail below in conjunction with the accompanying drawings, so that those skilled in the art can implement it with reference to the description.

[0048] Such as figure 1 As shown, the neural network-based machine self-learning provided by the present invention builds a knowledge map training method, including:

[0049] S100: Acquire the sentence sent by the user based on the natural scene, use the threshold speech noise reduction algorithm to filter and reduce the noise of the input sentence, and obtain the category of the sentence, and obtain the sentence above the sentence, and the sentence above the sentence category;

[0050] S200: Determine a matching feedback sentence according to the sentence category described in the sentence;

[0051] S300: If it does not exist, give an answer to the statement sent by the user according to the neural network dialogue model; including:

[0052] S310: The coding layer of 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 a machine self-learning construction knowledge atlas training method based on a neural network. The method comprises the following steps of acquiring a statement based on a natural situation which is sent by a user, using a voice noise reduction algorithm to carry out filtering noise reduction on an input statement, and determining a matched feedback statement; if the matched feedback statement does not exist, according to a neural network dialogue model, giving an answer of the statement sent by the user, which includes the following steps of constructing a coding layer of a user sending statement model to be a first neural network; analyzing a user sending statement in the first neural network and acquiring a first intermediate vector used for expressing a user sending statement meaning; and constructing a decoding layer of a dialogue generation model to be a second neural network, analyzing the intermediate vector in the second neural network, and acquiring a vector group used for expressing a statement answer. In the invention, a threshold voice noise reduction algorithm is used to acquire a small mean square error and a signal to noise ratio of a reconstruction voice signal is increased.

Description

technical field [0001] The invention relates to the field of intelligent robots, in particular to a neural network-based machine self-learning method for constructing a knowledge map training method. Background technique [0002] Chatterbot (chatterbot) is a program used to simulate human dialogue or chat. The reason for chatbots is that the developers put the answers they are interested in into the database. When a question is thrown to the chatbot, it uses the similarity matching algorithm to find the most similar question from the database, and then based on the question and The corresponding relationship of the answer, give the most appropriate answer, and reply to its chat partner. [0003] However, in the current robot chat scenario, when the same or similar question matching the user's requested question cannot be found in the robot knowledge base, the robot cannot return the correct or appropriate answer to the user. [0004] In addition to the problem of limited k...

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): G10L15/06G10L15/183G10L17/22G10L21/0208G10L25/30G10L25/72
CPCG10L15/063G10L15/183G10L17/22G10L21/0208G10L25/30G10L25/72
Inventor 刘颖博王东亮王洪斌
Owner 吉林省盛创科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products