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

False review detection method based on double cycle graph

A detection method and double-loop technology, applied in the field of network security, can solve the problems of poor filtering effect and low credibility of the initial value of confidence.

Active Publication Date: 2020-12-11
ZHEJIANG UNIV OF TECH
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, due to fewer considerations, the graph-based filtering algorithm has the problems of low confidence in the initial value of confidence and poor filtering effect.

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 review detection method based on double cycle graph
  • False review detection method based on double cycle graph
  • False review detection method based on double cycle graph

Examples

Experimental program
Comparison scheme
Effect test

experiment example

[0093] The three Yelp data sets used in the experiment example are:

[0094] YelpChi dataset, containing 67395 reviews, 38063 users and 201 stores;

[0095] YelpNYC dataset, containing 359,052 reviews, 160,225 users and 923 stores;

[0096] The YelpZip dataset contains 608,598 reviews, 260,277 users, and 5,044 stores.

[0097] Using the above false review detection method to test three Yelp data sets, the test results are:

[0098] In the YelpChi dataset, fake users account for about 20.33% of the total number of users, and fake reviews account for about 13.23% of the total number of reviews; in the YelpNYC dataset, fake users account for about 17.79% of the total number of users, and fake reviews account for about 10.27% of the total number of reviews; fake users in the YelpZip dataset Accounting for about 23.91% of the total number of users, fake reviews accounted for about 13.22% of the total number of reviews.

[0099] It can be seen that the false comment detection met...

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 false comment detection method based on a double-cycle graph, comprising the following steps: (1) using an original graph filter to calculate a comment confidence degree anda user confidence degree of the original comment data, and screening the user confidence degree to obtain a reliable user; (2) calculating the store confidence level of the comment data correspondingto the reliable users by using the original graph filter; (3) updating the comment confidence level in the original graph filter to the comment confidence level obtained in the step (1), and calculating the user confidence level of the original comment data by using the original graph filter; (4) taking the store confidence obtained in the step (2) and the user confidence obtained in the step (3)as initial values to construct a weighted map filter; (5) calculating The store confidence, user confidence and comment confidence of the original comment data by using the weighted graph filter, andthe false comment is obtained by screening according to the comment confidence. This method improves the detection accuracy of false comments.

Description

technical field [0001] The invention belongs to the technical field of network security, and in particular relates to a false comment detection method based on a double cycle graph. Background technique [0002] Most e-commerce allows users to comment on their services and quality online, and online reviews have gradually become the basis for consumers to shop, and users' online reviews of stores will greatly affect the store's reputation and sales. User reviews play an important role in recommendation systems. A large amount of real and effective review data can enable the recommendation system to generate effective recommendations, thereby providing consumers with the correct store or product recommendations. [0003] Driven by interests, some users will deliberately write reviews that do not match the reality, that is, fake reviews, with the purpose of exaggerating or defaming a certain store to maximize its benefits. A phenomenon in which a large number of fake reviews ...

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 Patents(China)
IPC IPC(8): G06F16/2458G06Q30/02
CPCG06Q30/0282
Inventor 陈晋音黄国瀚吴洋洋
Owner ZHEJIANG UNIV OF TECH
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