A method for identifying e-commerce trolls based on range

A recognition method, a water army technology, applied in the information field, can solve problems such as poor performance and poor robustness, and achieve the effects of high precision of results, easy understanding of the method, and small calculation space occupation

Active Publication Date: 2021-04-13
广州拉拉米信息科技股份有限公司
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

After relevant experiments, the above algorithm does not perform well when the amount of data is large and relatively sparse, and the robustness is not very good

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
  • A method for identifying e-commerce trolls based on range
  • A method for identifying e-commerce trolls based on range
  • A method for identifying e-commerce trolls based on range

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0060] The technical solutions and beneficial effects of the present invention will be described in detail below in conjunction with the accompanying drawings.

[0061] Such as figure 1 As shown, the present invention provides a method for identifying e-commerce trolls based on extreme difference, the idea is as follows:

[0062] (1) Introduce a "triple group" array to store user rating data. In order to solve the problem of big data storage in the e-commerce system, a triple data structure G={i,j,k} is defined to represent users, products and ratings respectively. The initialization of a triplet data structure represents a scoring behavior, and the establishment of a triplet array can cover all scoring behaviors;

[0063] (2) Initialize the product quality. Each product has an inherent product quality according to its pros and cons, and the product quality is formed according to the user's evaluation in the scoring system. However, due to the existence of malicious raters,...

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 method for identifying e-commerce trolls based on extreme difference. The steps are: calculating the commodity quality in the initial state, calculating the deviation between each user's rating and the commodity quality; calculating the scoring extreme difference of each user; based on the deviation and the scoring extreme difference, calculate the user reputation of each user; substitute each user reputation into the product quality calculation formula to obtain the corresponding product quality, and then get the new user reputation; calculate the total change of user reputation, if it is greater than the reputation change threshold, the new Substitute the user reputation into the product quality calculation formula to obtain the corresponding product quality until the sum of user reputation changes is less than the reputation change threshold; if it is less than the reputation change threshold, stop the iteration; finally, select the top N low reputation users as trolls. This method has a better ability to identify extreme water army groups and random water army groups, and can basically guarantee the original recognition ability under big data, with strong robustness and strong scalability.

Description

technical field [0001] The invention belongs to the field of information technology, relates to complex network analysis technology, and relates to a reputation evaluation method based on information entropy and user rating features in an e-commerce system, specifically a method for evaluating a weighted user-commodity bipartite graph. The calculation method of user reputation evaluation. Background technique [0002] With the rapid development of e-commerce, a large number of commodity transactions are increasingly dependent on a reliable reputation system to give reasonable and valuable ratings to items. However, the current reputation scoring system is facing many challenges, the most common of which is that users The unreasonable problems caused by the arbitrariness of scoring, especially the organized troll groups deliberately increase or decrease the score of specific merchants, seriously mislead consumers in making decisions. These troll groups do not score based on ...

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): G06Q30/06G06Q30/02
CPCG06Q30/0282G06Q30/0601G06Q30/0609
Inventor 孙宏亮梁楷平卜湛曹杰
Owner 广州拉拉米信息科技股份有限公司
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