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

Conversation interaction method, device and equipment and storage medium

An interactive method and sentence technology, applied in semantic analysis, natural language translation, instruments, etc., can solve the problems of lack of dialogue context logic, low accuracy of multiple rounds of dialogue, high cost of training corpus, etc., to ensure diversity and ensure diversity performance, and the effect of improving accuracy

Active Publication Date: 2020-06-09
BAIDU ONLINE NETWORK TECH (BEIJIBG) CO LTD +1
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the existing technology is usually only able to achieve the effect of a single round of dialogue, lacking the logic of the dialogue context, the accuracy of multiple rounds of dialogue is low, and the cost of training corpus is high

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
  • Conversation interaction method, device and equipment and storage medium
  • Conversation interaction method, device and equipment and storage medium
  • Conversation interaction method, device and equipment and storage medium

Examples

Experimental program
Comparison scheme
Effect test

no. 1 example

[0062] figure 1 It is a flow chart of a dialog interaction method according to the first embodiment of the present application. This embodiment is applicable to the situation of determining response sentences in multiple rounds of dialog between man and machine, and the method can be executed by a dialog interaction device. The device is realized by means of software and / or hardware, and is preferably configured in an electronic device, such as a smart speaker. Such as figure 1 As shown, the method specifically includes the following:

[0063] S110. Determine the first semantic code of the received user information according to the sentence tree.

[0064] In the specific embodiment of the present application, the user information refers to the dialog voice input by the user to the electronic device, which may be the initial voice initiated by the user to initiate a dialog to the electronic device after the user wakes up the electronic device, or may be the user's voice durin...

no. 2 example

[0082] figure 2 It is the training flowchart of the sentence tree and the dialogue tree according to the second embodiment of the present application. On the basis of the above-mentioned first embodiment, this embodiment further explains the training methods of the sentence tree and the dialogue tree, which can be based on the logical brain Sentence node information and / or word node information in graph samples are trained. Such as figure 2 As shown, the method specifically includes the following:

[0083] S210. According to the logical brain map sample, learn to obtain the nodes in the sentence tree and the dialogue tree.

[0084] In the specific embodiment of the present application, the logical brain map refers to expressing the words and / or sentences involved in the dialogue through the hierarchical relationship of mutual affiliation and correlation nodes. Logical brain maps can be constructed for different scenarios in advance to represent the dialogue logic in diffe...

no. 3 example

[0114] Image 6 It is a flow chart of a dialogue interaction method according to the third embodiment of the present application. On the basis of the above-mentioned first embodiment, this embodiment further explains the method of determining the target response sentence, and can use the words in the sentence tree Node traversal is used to identify the semantics of user information, and the response semantics are determined through the response relationship between sentence nodes in the dialogue tree, and finally the target response sentence is determined through the extraction of word nodes under the response semantics in the sentence tree. Such as Image 6 As shown, the method specifically includes the following:

[0115] S610. According to the word information of the user information, traverse from the word node to the sentence node of the sentence tree, and determine a target word node combination constituting the user information.

[0116] In the specific embodiment of ...

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 discloses a conversation interaction method and device, equipment and a storage medium, and relates to the technical field of artificial intelligence. According to thespecific implementation scheme, the method comprises the steps of determining a first semantic code of received user information according to a sentence tree; determining a second semantic code for responding to the user information from a conversation tree according to the first semantic code; wherein the sentence tree and the conversation tree are obtained by training sentence node information and / or word node information in a logic brain graph sample; and determining a target response sentence of the second semantic code from the sentence tree, wherein the target response sentence is used for conversation with the user. According to the embodiment of the invention, the conversation logic is quickly learned into the sentence tree and the conversation tree on the basis of the logic braindiagram, and through alternate matching between the sentence tree and the conversation tree, the response content considers the conversation content and contains multiple representations of the conversation content, so that the diversity of the conversation content is guaranteed, and the accuracy of multiple rounds of conversations between human and machines is improved.

Description

technical field [0001] The embodiments of the present application relate to the field of computer technology, in particular to the field of artificial intelligence technology, and specifically to a dialog interaction method, device, device, and storage medium. Background technique [0002] With the popularity of smart devices such as smart speakers, the demand for dialogue between users and smart devices is increasing. In addition to resource on-demand, home control and other needs, most users also have emotional exchanges, daily conversations, and topic chats with smart devices. and other human-computer interaction behavior. However, the existing technology is usually only able to achieve the effect of a single round of dialogue, lacking the logic of the dialogue context, the accuracy of multiple rounds of dialogue is low, and the cost of training corpus is high. Contents of the invention [0003] Embodiments of the present application provide a dialog interaction method...

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): G06F16/36G06F16/332G06F40/30G06F40/205
CPCG06F16/367G06F16/3329G06F40/242G06F40/289G06F40/35G06F40/56
Inventor 贺文嵩苗亚飞汤其超徐犇谢剑
Owner BAIDU ONLINE NETWORK TECH (BEIJIBG) CO LTD
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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com