Unlock instant, AI-driven research and patent intelligence for your innovation.

Early public opinion detection method based on deep learning

A technology of deep learning and detection methods, applied in the fields of unstructured text data retrieval, special data processing applications, instruments, etc., which can solve the problem of comment dependence, low detection accuracy of suspected rumor information, and inability to construct a rumor propagation structure, etc. problem, to solve the comment dependency problem, enhance interpretability and robustness

Pending Publication Date: 2021-10-01
浙江华巽科技有限公司
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In the case of less comment data, this kind of method cannot construct the spread structure of rumors, and the detection accuracy of suspected rumor information is not high, so there is comment dependence, and it is not suitable for the early detection task of rumors

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
  • Early public opinion detection method based on deep learning
  • Early public opinion detection method based on deep learning
  • Early public opinion detection method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0017] Embodiment 1: overall framework of the present invention and concrete respectively as figure 1 and figure 2 As shown, an early public opinion detection method based on deep learning, the specific implementation steps are as follows:`

[0018] Step 1, use the GADT model to obtain a deep representation of the text. For rumor posts, the text and the corresponding dependency tree are used as input. First, BERT is used to obtain the word vector representation, and then the dependency relationship between words is obtained through syntactic analysis. The text content of the entire post is regarded as a graph structure, and finally GAT is used. The network performs further enhanced embeddings. The model encodes and embeds text information through the context and dependencies between words, and provides a deep text representation for rumor detection tasks. The implementation process of this step is divided into 3 sub-steps:

[0019] Sub-step 1-1, use BERT to obtain the wor...

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 an early public opinion detection method based on deep learning, and provides a rumor detection method based on text and user characteristics, aiming at the problem of comment dependency in early detection of related public opinions of a rumor, namely ERD-GAT-VAE (Early Rumor Detection Based On Garph Attention Network and Variable Auto-encoder). Firstly, on the premise of not depending on comment data, the invention provides a graph attention over dependency tree model (GADT) for acquiring a deep representation vector of a rumor text; secondly, the invention provides a user credibility evaluation model (UCE) based on a variational auto-encoder for obtaining a user credibility representation vector; finally, an early detection result of rumor-related public opinions is obtained by fusing the text deep representation and the user credibility representation vector.

Description

technical field [0001] The invention relates to an early public opinion detection method based on deep learning, which belongs to the technical field of Internet and artificial intelligence. Background technique [0002] In recent years, social networks have achieved rapid development, and have quickly become one of the important ways for people to obtain news information. Due to the huge amount of information and the speed of transmission far faster than traditional media, a large number of unconfirmed rumors can be spread freely in cyberspace, which has become a growing problem. In the early stage of the development of public opinion events, the false information contained in social network rumors was highly misleading and inflammatory, causing adverse social influences and threatening social harmony and stability. Therefore, rapid detection of rumors during their early dissemination is an important task for early public opinion detection. [0003] Most of the existing r...

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): G06F16/35G06F40/126G06F40/14G06F40/211
CPCG06F16/35G06F40/126G06F40/14G06F40/211
Inventor 杨鹏匡晨冷俊成刘子健
Owner 浙江华巽科技有限公司