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

Deep attention rumor identification method and device based on ternary features

A technology of attention and attention model, applied in special data processing applications, instruments, text database query, etc., can solve the problems of inaccurate identification and single use of features, and achieve the effect of overcoming single features and improving accuracy.

Active Publication Date: 2020-04-14
WUHAN UNIV
View PDF7 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] In view of this, the present invention provides a deep attention rumor identification method and device based on ternary features to solve or at least partly solve the problem of the end-to-end method on features in the prior art due to the high requirements for expert knowledge. Utilization is too single, which leads to technical problems of inaccurate identification

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 attention rumor identification method and device based on ternary features
  • Deep attention rumor identification method and device based on ternary features
  • Deep attention rumor identification method and device based on ternary features

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0064] This embodiment provides a deep attention rumor identification method based on triple features, please refer to figure 1 , the method includes:

[0065] Step S1: Extract the propagation track during the information dissemination process from the given data set, use the network topology structure to represent the propagation track during the propagation process, and express the network topology structure with segment vectors.

[0066] Step S2: Extract the text topic of the information within the preset time period from the given data set, and represent the text topic of the information with a segment vector;

[0067] Step S3: extract the user's feedback signal from the given data set, and construct a feedback signal vector.

[0068] Specifically, due to the temporal variability of the social network itself, in the traditional methods for identifying rumors on social networks, the model is only constructed for the overall data characteristics or the original data content...

Embodiment 2

[0152] Based on the same inventive concept, this embodiment provides a deep attention rumor identification device based on ternary features, please refer to Figure 7 , the device consists of:

[0153] The propagation trajectory extraction module is used to extract the propagation trajectory in the process of information dissemination from a given data set, using the network topology to represent the propagation trajectory in the propagation process, and expressing the network topology with a segment vector;

[0154] The text topic extraction module is used to extract the text topic of the information within the preset time period from the given data set, and express the text topic of the information with a segment vector;

[0155] The user feedback signal extraction module is used to extract the user's feedback signal from a given data set, and construct a feedback signal vector;

[0156] The attention module is a building block, which is used to use the segment vector of th...

Embodiment 3

[0161] Based on the same inventive concept, the present application also provides a computer-readable storage medium 300, please refer to Figure 8 , on which a computer program 311 is stored, and when the program is executed, the method as described in the first embodiment is realized.

[0162] Since the computer-readable storage medium introduced in the third embodiment of the present invention is the readable storage medium used to implement the deep attention rumor identification method based on the ternary feature in the first embodiment of the present invention, it is based on the introduction in the first embodiment of the present invention Those skilled in the art can understand the specific structure and deformation of the computer-readable storage medium, so details will not be repeated here. All computer-readable storage media used in the method in Embodiment 1 of the present invention fall within the scope of protection intended by the present invention.

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 attention rumor identification method and device based on ternary features. A public social network platform rumor identification data set is adopted to extract text content of the information in the time period, propagation trajectory of events, the feedback signal of the user as the ternary features, and an improved biased random walk algorithm based on space-timestructure similarity and node measurement is provided for distributed representation learning of the nodes; an improved vectorization method of a propagation network node and an explicit vector representation method of a network topology structure are used for mapping the network topology structure from high dimension to low dimension representation, a time sequence identification method based onself-attention is adopted, and the feasibility of the method on a real world data set is verified. According to the method, rumor information can be effectively identified and the identification accuracy is improved.

Description

technical field [0001] The invention relates to the technical field of text processing and detection, in particular to a method and device for identifying rumors with deep attention based on ternary features. Background technique [0002] In the current era, social media platforms have vigorously promoted the explosive dissemination of information on various events because of their convenience, and have become an important source for ordinary people in today's society to obtain external information. However, the development of social networks and the increase in the amount of information have also brought about the proliferation of various types of bad information such as rumors. At the same time, because the amount of information on social media is too large, it is time-consuming and labor-intensive to manually dispel rumors, which is not ideal, so the need for automated rumor detection is imminent. [0003] The existing methods can be divided into two categories according...

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): G06F16/9536G06F16/33
CPCG06F16/9536G06F16/3344
Inventor 王丽娜王文琦柯剑鹏叶傲霜陈铜唐奔宵
Owner WUHAN 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