Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

False message early detection method based on crowd-sourcing data fusion

A data fusion and early detection technology, applied in the field of deep learning recognition systems, can solve the problem that the recognition method is easy to be cracked in a targeted manner, and achieve the effect of avoiding the process of manually setting features and improving the accuracy.

Pending Publication Date: 2020-02-07
NORTHWESTERN POLYTECHNICAL UNIV
View PDF3 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although these methods have good accuracy in identifying false news, the content-based identification method is easy to be cracked, and the other two methods require sufficient data support

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
  • False message early detection method based on crowd-sourcing data fusion
  • False message early detection method based on crowd-sourcing data fusion
  • False message early detection method based on crowd-sourcing data fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0021] The technical solution of the present invention will be further described below in conjunction with the drawings: figure 1 , figure 2 Shown.

[0022] Step 1. Obtain the posts of news events on the social media platform, as well as the comments under the posts and relevant information of the users who made the comments.

[0023] Step 2: Mark the authenticity of news events and judge them on the Sina Weibo anti-rumor platform, Snopes.com, Politifact.org, etc.

[0024] Step 3: Define a news event message E={P,S}, with the original text P of the news event, and a time series S={s 1 , S 2 ,...,S n }, where s i ={u i , T i , C i } Contains user u i At time t i Comments left c i ,Such as figure 2 Shown. Where u i ={a 1 , A 2 ,..., a n }, the goal is to obtain a prediction function F(s) such that it satisfies:

[0025]

[0026] Step 4: First, arrange the comments under the news event in time sequence, divide the comments into segments, use the TF-IDF algorithm between the segments to...

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 provides a false message early detection method based on crowd-sourcing data fusion. The method comprises the following steps: establishing a comment model; learning comment features byusing Attention-RNN of two layers of GRU units; establishing a user attribute model; learning user attribute features by using the CNN; and establishing a joint model: fusing the two parts of featuresto obtain a final message identification result. The method comprises the following steps: learning a time sequence and content characteristics of comments under a message through an Attention-RNN (Recurrent Neural Network); using a CNN (Convolutional Neural Network) for learning attribute features of a comment user under a message, finally, fusing the two parts of features, and obtaining a recognition result through a full-connection neural network. Early scarce data resources are fully utilized, a heavy link of manually setting features is avoided, and the recognition accuracy is improved.

Description

Technical field [0001] The invention relates to the field of deep learning recognition systems, and in particular to a false message early detection method based on group intelligence data fusion. Background technique [0002] Early identification of fake news on social media platforms is one of the necessary security applications in the Internet age. At present, there have been a large number of researches on the identification of fake messages. They are based on the content, dissemination, user and other perspectives to identify the authenticity of the message. Specifically, content-based fake news recognition is similar to the study of fake news from the perspective of knowledge or rules, and judgments are mainly based on specific writing styles or sensational titles in fake news, such as vocabulary features, syntactic features, and topics Features, image features, etc. Based on the dissemination of false news identification, this type of method mainly uses the dissemination...

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): G06F16/21G06N3/04
CPCG06F16/212G06N3/045
Inventor 郭斌吴广智丁亚三於志文王柱王亮梁韵基
Owner NORTHWESTERN POLYTECHNICAL UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
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