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

Automatic reply generation method for open domain dialogue system

A dialogue system and automatic generation technology, applied in the field of human-computer dialogue of natural language processing, can solve the problem of lack of semantic information in the reply

Inactive Publication Date: 2019-10-15
UNIV OF ELECTRONIC SCI & TECH OF CHINA
View PDF3 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The reply generation method based on the sequence model finally uses the maximum likelihood estimation method to determine the generated words, which can easily lead to the selection of words with high frequency in the training data, so that the generated reply is a "universal reply" lacking semantic information.

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
  • Automatic reply generation method for open domain dialogue system
  • Automatic reply generation method for open domain dialogue system
  • Automatic reply generation method for open domain dialogue system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0085] In order to make the purpose, technical solutions and advantages of the present invention clearer, the technical solutions will be clearly and completely described below in conjunction with the embodiments.

[0086] Step 1: Build a dialogue corpus. Each sample in the corpus is composed of a user input sentence and a user reply sentence, recorded as {(X 1 ,Y 1 ),(X 2 ,Y 2 ),…,(X N ,Y N )}; in this embodiment, assume that there is a training sample in the dialogue corpus: (X, Y)=("Good morning!", "Good morning, have you had breakfast?"), wherein, "Good morning!" is The input sentence, "Good morning, have you had breakfast?" is a reply sentence; first, the input sentence X and the reply sentence Y are word-segmented, and the word sequence {"morning", "good", "!"} and the reply of the input sentence X are obtained The word sequence of statement Y {"early", ",", "eat", "breakfast", "ha", "do", "?");

[0087] Then use the common word vector table to convert each word in...

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 an automatic reply generation method for an open domain dialogue system, and aims to enable the dialogue system to automatically generate a reply statement which conforms to semantics and contains information quantity according to an input statement of a user. The method comprises the following steps: firstly, constructing a dialogue corpus, preprocessing input and reply statements of each sample to obtain word vector sequences of the input and reply statements, and respectively inputting the word vector sequences into two encoders to obtain semantic representation vectors of the input and reply statements; secondly, retrieving knowledge graphs containing entity words in a knowledge base according to the entity words contained in the input statement, and calculatinga vector of each knowledge graph; thirdly, fusing the knowledge graph vector with the semantic representation vectors of the input statement and the reply statement through two fusion networks, and inputting the fused vectors into a decoder to generate the reply statement; fourthly, training a reply automatic generation model based on a corpus; and finally, for a new input statement, utilizing the reply automatic generation model to generate a reply.

Description

technical field [0001] The invention belongs to the field of human-computer dialogue of natural language processing, and in particular relates to an automatic reply generation method of an open domain dialogue system based on a deep neural network. Background technique [0002] Human-computer dialogue systems can be divided into closed-domain dialogue systems and open-domain dialogue systems according to different application fields. Closed-domain dialogue systems usually have clear goals and limited knowledge scope, and only generate responses to domain-specific user input, for example, dialogue systems for airline ticket reservations and restaurant order dialogue systems. Open-domain dialogue systems usually have no clear goals and limited knowledge scope. Users and the system can have free dialogues on any topic, such as chatting robots, as chat partners, can freely chat with users. attached figure 1 A diagram showing a conversation between a person and a chatbot. The ...

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): G06F16/33G06F16/332
CPCG06F16/3329G06F16/3344
Inventor 刘梦娟孟园赵培包笑明刘瑶
Owner UNIV OF ELECTRONIC SCI & TECH OF CHINA
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