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

Deep learning-based natural language generation method

A natural language and deep learning technology, applied in natural language data processing, biological neural network models, special data processing applications, etc., can solve problems such as difficulty in application, solve generation problems, reduce the degree of manual participation, and hide state vectors. accurate effect

Inactive Publication Date: 2018-09-21
SHENZHEN GRADUATE SCHOOL TSINGHUA UNIV
View PDF2 Cites 19 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

But for some relatively free texts, this technique is difficult to apply

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
  • Deep learning-based natural language generation method
  • Deep learning-based natural language generation method
  • Deep learning-based natural language generation method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0014] The present invention will be described in further detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0015] Such as figure 1 As shown, the deep learning-based natural language generation device for generating news comments in this specific embodiment includes a comment generation device and a comment discrimination device. Among them, the review generation device includes four sub-modules: encoding module, decoding module, gate attention module and emotion control module. The comment generation means generates news comments corresponding to the input news documents. The comment discrimination device includes two sub-modules: a feature extraction module and a feature classification module. The comment discriminating device is used for judging the authenticity of the input comment, so as to distinguish the real comment from the generated comment, and obtain a feedback value for improving the comment generating devic...

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 deep learning-based natural language generation method. The method comprises the steps of training a comment generation apparatus by using existing news and comments. The training stage comprises the following steps of S1, performing vectorization processing on words in the existing news and comments to obtain word vectors corresponding to the words in the news and word vectors corresponding to the words in the comments; S2, obtaining hidden state vectors of the words in the news; S3, obtaining hidden state vectors of the words in the comments; S4, processing the words in the comments to obtain new hidden state vectors of the words; and S5, according to the obtained new hidden state vectors of the words in the comments in the step S4, predicting the next word corresponding to each word: according to the hidden state vector of the current word, obtaining a probability of selecting each word in a word table, and selecting the word with the maximum probability asthe next word of the current word in the generated comments. According to the method, the corresponding news comments can be generated for the given news, and the accuracy of word selection in the comments is relatively high.

Description

【Technical field】 [0001] The invention relates to the field of computer applications, in particular to a method for generating natural language based on deep learning. 【Background technique】 [0002] Natural language generation is an interdisciplinary subject of artificial intelligence and computational language, and its purpose is to enable machines to generate understandable human language texts. Advances in natural language generation technology can help build strong artificial intelligence systems and improve understanding of human language. Traditional natural language generation technology has been successfully applied in many fields, such as automatic news writing, which enables the machine to automatically generate sports event reports, and has the characteristics of fast and accurate. Tencent, Baidu, Toutiao and other companies have developed corresponding writing manuscript robot. There are also automatic weather forecast generation, automatic reply generation in...

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): G06F17/27G06F17/22G06N3/04
CPCG06F40/151G06F40/284G06N3/045
Inventor 王伟郑海涛陈金元韩金新肖喜
Owner SHENZHEN GRADUATE SCHOOL TSINGHUA UNIV
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