Rumor recognition method

A method for identifying rumors and a technology for rumors, applied in the fields of the Internet and artificial intelligence, can solve problems such as inability to identify rumors, achieve the advantages of strong timeliness, reduce computational complexity, and improve the degree of adaptation.

Inactive Publication Date: 2018-10-02
ZHONGAN INFORMATION TECH SERVICES CO LTD
View PDF6 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The technical problem to be solved by the present invention is to provide a rumor identification me

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
  • Rumor recognition method
  • Rumor recognition method
  • Rumor recognition method

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0066] Example two

[0067] Reference image 3 As shown, there are still some shortcomings in judging the text to be tested by the rumor discrimination model alone. In this embodiment, a user classification and weighting system is introduced, which is essentially based on users with specific marks (users of a certain level in a certain field). "Text to be tested" is scored for rumor determination.

[0068] The users in the user classification and weighting system include initial users and non-initial users.

[0069] (1) User classification initialization

[0070] New users can choose whether they are willing to become volunteers and select areas of interest or expertise through a simple process. This part of the selection is used as the initial classification of the user, and the initial weight of the user is 0.

[0071] (2) Initialize user weight adjustment

[0072] Because there may be differences between the user's self-assessment and the objective facts, we need to evaluate the use...

Example Embodiment

[0082] Example three

[0083] The rumor recognition feedback model based on human-computer collaboration uses the "user weighted judgment" as the standard to compare the scores of the "rumor discrimination model", save the wrong annotation data, and iterate into the model.

[0084] In this embodiment, the user accesses the rumor discrimination model in two ways. One is to access the API interface and call the predict function shown in the "model effect test" code in the rumor discrimination model (5). The content of the evaluation is input as a variable; the other is through the test page, and the text to be tested is input into the page.

[0085] After collecting the rumor text to be tested entered by the user, it can be a link or text data, and then the prediction function in the rumor judgment model is also called to obtain the judgment result. At the same time, both the pre-evaluation result and the text content are recorded in the database.

[0086] After the text is classified i...

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 present invention discloses a rumor recognition method. The method comprises: grabbing marked text content, cleaning and sorting the text, dividing the text into sentences, and storing the text content in sentences into a database; according to the Chinese word segmentation vocabulary, performing word segmentation on the text content which is segmented into sentences, sorting the word segmented content into a model readable and standardized form, and taking the model readable and standardized form as the input content of the model; performing high latitude vectorization association mappingon the word segmented content through the model, performing the weighted combination on word vectors to obtain a sentence vector, a paragraph vector or an article vector, and outputting the vectors from the model; inputting the processed training corpus information to a classifier model for training, and outputting a determination result; and constructing a to-be-detected text function, and performing an effect test on a rumor determination model. According to the method disclosed by the present invention, while reducing the labor cost, the reliability of rumor determination is improved.

Description

technical field [0001] The invention belongs to the field of Internet and artificial intelligence, and in particular relates to a rumor identification method. Background technique [0002] In today's era of Internet information explosion, it is very easy for people to obtain all kinds of information, but when faced with massive information, people often lack the ability to distinguish things themselves, and will be caught by some professional, disseminated, and fabricated information. People are confused by speech, so as to make wrong judgments, which are even unfavorable to social development. This kind of speech is a well-known rumor. Rumors can be roughly divided into six categories based on the field of content: food safety, medical health, science and technology, legends and anecdotes, pet flowers, and natural environment. Rumors can harm individuals, groups, and even society, making people's simple and stable interpersonal relationships complicated and tense, and maki...

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): G06F17/30G06F17/27
CPCG06F40/216G06F40/289
Inventor 陈鸿睿肖日新马斌纪其进
Owner ZHONGAN INFORMATION TECH SERVICES CO LTD
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
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap