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

Text generation method guided by different facet viewpoint evolution trends

A faceted and point-of-view technology, applied in unstructured text data retrieval, text database clustering/classification, biological neural network model, etc., to achieve more suitable and targeted effects

Active Publication Date: 2020-06-02
XI AN JIAOTONG UNIV
View PDF3 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] The object of the present invention is to provide a kind of text generation method guided by the evolution trend of different facet viewpoints, to solve the above problems

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
  • Text generation method guided by different facet viewpoint evolution trends
  • Text generation method guided by different facet viewpoint evolution trends
  • Text generation method guided by different facet viewpoint evolution trends

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0060] Taking the corpus of "the bus falling into the river incident" on Weibo as an example to illustrate the text generation process guided by the evolution trend of multi-faceted public opinion opinions.

[0061] Such as figure 1 The process of constructing facet classifier and opinion classifier is shown in the figure.

[0062] First, collect all the blog posts of "the bus falling into the river incident" on Weibo as the source corpus. After cleaning the source corpus, divide the blog posts into multiple clauses based on punctuation marks.

[0063] According to expert knowledge or public opinion event management needs, determine the three important facets of the public opinion event—the bus driver, the female driver, and the beating female passenger. For each facet, a keyword or phrase description is given, that is, facet description .

[0064] Use the source corpus to train the word2vec word vector model, so as to obtain the word embedding vector representation of the c...

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 text generation method guided by different facet viewpoint evolution trends, and the method comprises the steps: (1) constructing a facet classifier and a viewpoint classifier: collecting all blog articles of a specific hot event in a social network as a source corpus set, determining an important facet of a public opinion event according to expert knowledge or public opinion event management demands, and giving out facet description; realizing clause extraction of different facets in the source corpus set according to semantic distances between the corpus clauses andfacet description so as to train a facet classifier; clustering corpora of different viewpoints in each facet by using a text clustering method based on viewpoints, and training a viewpoint classifier by using the clustered corpora; (2) generating an adversarial viewpoint text; wherein the text generation model comprises a generator and three discriminators for adversarial training; wherein the generator generates a text which has the same facet as the corpus sentence and has opinion adversarial, the three discriminators distinguish the sentence generated by the generator from the source corpus sentence, and the adversarial training is repeated until the three discriminators cannot distinguish the sentence generated by the generator from the source corpus sentence.

Description

technical field [0001] The invention belongs to the field of anti-guiding text generation of opinion evolution trends in multi-faceted public opinion events, and in particular relates to a text generation method guided by different facet opinion evolution trends. Background technique [0002] Nowadays, various self-media platforms are developing rapidly, and the Internet has become the main place for people to obtain information, participate in public affairs, and express their personal opinions. However, due to the characteristics of anonymity and fast dissemination of the Internet, all kinds of false and negative information and speeches flood the Internet, especially for social events related to the law, people's livelihood, and public morality. Maliciously inciting netizens' emotions and spreading false information will seriously endanger society. Stablize. Internet public opinion has become an important factor affecting social stability. On the basis of establishing co...

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): G06F16/35G06N3/04
CPCG06F16/35G06N3/044G06N3/045Y02D10/00
Inventor 秦涛王熙凤郑庆华罗敏楠张玉哲王伟帆王浩年
Owner XI AN JIAOTONG 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