Unlock instant, AI-driven research and patent intelligence for your innovation.

Multi-true-value conflict resolution method based on Bayesian model

A Bayesian model, conflict resolution technology, applied in character and pattern recognition, special data processing applications, instruments, etc., can solve problems such as weak research, distortion of malicious data sources, etc., to achieve the effect of improving efficiency and accuracy

Active Publication Date: 2020-09-25
XI AN JIAOTONG UNIV
View PDF4 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The truth value largely depends on the reliability of the data source, and the truth inference results are easily distorted by malicious data sources
Therefore, the current research on multiple truth discovery is still very weak

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
  • Multi-true-value conflict resolution method based on Bayesian model
  • Multi-true-value conflict resolution method based on Bayesian model
  • Multi-true-value conflict resolution method based on Bayesian model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0054] The invention provides a multi-truth value conflict resolution method based on the Bayesian model. By identifying the deficiency of the unified initialization of the data source, the prior credibility evaluation of the data source is performed before the truth value is inferred, and then the value is substituted into the truth value. In value inference, the efficiency and accuracy of subsequent truth value inference can be improved, and the number of iterations can be reduced. In the multi-truth value inference, two factors, the similarity between the data source and the object and the confidence score of the multi-true value, are added to the true value. In the inference model, the true value is obtained; the credibility of the data source can also be calculated through the distribution of the true value. The frame diagram of multi-truth conflict resolution is as follows: figure 1 shown.

[0055] The present invention is achieved through the following technical soluti...

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 multi-true-value conflict resolution method based on a Bayesian model. The method comprises the following steps: analyzing book and movie data, generating a basic data set Draw, converting the basic data set Draw into a fact data set Df again, further generating a declaration data set Dc, and carrying out data source prior credibility evaluation by adopting a method basedon Gibbs sampling; obtaining data source prior reliability, the data source priori value is used as an initialization value of each data source, different professional knowledge degrees are displayedin different fields based on the data sources, the data richness in one field is high, richness measurement is converted into distance-based measurement, and the similarity between the data sources and objects providing data is measured; in the final Bayesian inference, the similarity between the data source and the object and the confidence coefficient of the truth value are considered as factors influencing truth value inference, and finally a truth value data set Dt is output to perform data source prior credibility estimation, so that the truth value inference efficiency and accuracy canbe improved.

Description

technical field [0001] The invention belongs to the technical field of data mining, and in particular relates to a multi-truth conflict resolution method based on a Bayesian model. Background technique [0002] In the information age, attribute values ​​of the same entity provided by different data sources may conflict. Therefore, identifying or generating correct answers from these potentially conflicting information is a non-trivial task. The most important challenges of this task are estimating the reliability of data sources and selecting high-quality answers. Currently, one of the simplest and most straightforward strategies is to conduct a majority vote, selecting the answer provided the most as the correct answer. However, this simple strategy does not take into account the reliability differences of data sources. Correct data can be overshadowed by false data provided by a bunch of malicious data sources. Also, this strategy tends to provide the most "trustworthy...

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): G06F16/2458G06K9/62
CPCG06F16/2465G06F18/22
Inventor 魏笔凡邓婷刘均郑庆华卜德蕊郝子琛郭敏
Owner XI AN JIAOTONG UNIV