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

End-to-end target guiding type conversation method based on deep learning

A goal-guided and deep learning technology, applied in the field of end-to-end goal-guided dialogue based on deep learning, can solve problems such as low dialogue efficiency and low accuracy, and achieve the effect of improving dialogue efficiency and enhancing personalized experience.

Active Publication Date: 2020-03-03
GUANGDONG UNIV OF TECH
View PDF2 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The present invention overcomes the technical defects of low dialogue efficiency and low accuracy of the existing end-to-end dialogue question-and-answer mode, provides an end-to-end goal-guided dialogue method and system based on deep learning, and proposes a knowledge base The query mechanism of specifying rows first and then specifying columns

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
  • End-to-end target guiding type conversation method based on deep learning
  • End-to-end target guiding type conversation method based on deep learning
  • End-to-end target guiding type conversation method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0031] see figure 1 , an end-to-end goal-guided dialogue method based on deep learning, including the following steps:

[0032] S1: Obtain the last round of user dialogue and the current dialogue history, and initialize the sequence-to-sequence model according to the previous round of user dialogue and the current dialogue history;

[0033] S2: Determine the line number of the current knowledge base entity;

[0034] S3: Determine the column number of the current knowledge base entity;

[0035] S4: Obtain the best matching entity through the attention mechanism;

[0036] S5: Steps S2-S4 are iteratively executed until the next round of dialogue is finally output.

[0037] The sequence-to-sequence model includes an encoder and a decoder. After the encoder is given the dialogue history and the user's last round of dialogue, the encoder encodes and abstracts the output content, obtains the last context vector, and transmits the context vector to the decoder. ; The decoder recei...

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 the technical field of natural language processing, in particular to an end-to-end target guide type conversation method based on deep learning, which comprises the followingsteps: S1, acquiring a previous round of user conversation and a current conversation history, and initializing a sequence to a sequence model according to the previous round of user conversation andthe current conversation history; S2, determining a row number of a current knowledge base entity; S3, determining a column number of the current knowledge base entity; S4, obtaining an optimal matching entity through an attention mechanism; and S5, iteratively executing the steps S2-S4 until the next round of conversation is finally output. According to the invention, the technical defects of lowend-to-end conversation efficiency and influence on personalized experience of users in the prior art are overcome, and good user service is provided through the accuracy of natural language conversation.

Description

technical field [0001] The invention relates to the technical field of natural language processing, in particular to an end-to-end goal-guided dialogue method based on deep learning. Background technique [0002] With the development of deep learning in recent years, the effect of the end-to-end goal-guided dialogue system has gradually improved. In our work, it is also an end-to-end dialogue system model. In the end-to-end model, the input user's dialogue is directly The system's reply can be output without going through the display conversion between different modules. [0003] However, the existing multi-round dialogue tasks in a specific field mainly use the sequence-to-sequence model. The accuracy of the dialogue generated according to the sequence-to-sequence model is low, and it cannot give users accurate and reliable answers. For multi-round dialogue tasks in a specific field, an important measure is whether the system can reply accurate answers with knowledge entit...

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): G06F16/332G06F16/35
Inventor 叶志豪廖朝辉蔡瑞初崔洪刚张袁震宇
Owner GUANGDONG UNIV OF TECH
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