A biased crowd-sourced information inference method based on dynamic bayesian game

By constructing a dynamic Bayesian game model, the problem of accurately inferring strategically biased user feedback in crowdsourcing platforms is solved, achieving efficient and accurate estimation of service status, applicable to scenarios such as online service evaluation and real-time traffic reporting.

CN122198155APending Publication Date: 2026-06-12NORTHWESTERN POLYTECHNICAL UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
NORTHWESTERN POLYTECHNICAL UNIV
Filing Date
2026-05-15
Publication Date
2026-06-12

AI Technical Summary

⚠Technical Problem

Existing technologies make it difficult to accurately infer service status from strategically biased user feedback in crowdsourcing platforms, leading to reduced credibility of evaluation systems and low efficiency in information utilization.

⚗Method used

A two-stage dynamic Bayesian game model between users and the platform is constructed, defining the user's private bias type and utility function. By using a unified Bayesian inference rule, true information is extracted from mixed messages, thus achieving accurate estimation of service status.

🎯Benefits of technology

As the number of users increases, the inference accuracy approaches the theoretical optimum, outperforming traditional methods. This improves information utilization efficiency and inference accuracy, making it suitable for online service evaluation, real-time traffic reporting, and network quality monitoring.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122198155A_ABST
    Figure CN122198155A_ABST
Patent Text Reader

Abstract

The application discloses a biased crowd-sourcing information inference method based on dynamic Bayesian game, a two-stage dynamic Bayesian game model of a user and a platform is constructed, and a perfect Bayesian equilibrium is solved, the user generates a message according to a private biased type and an observed service type signal and sends the message to the platform, the platform counts the number of high type messages in the received message, calculates a posterior probability distribution of a service state type based on a unified Bayesian inference rule, and then outputs an optimal inference action as a service state estimation. Under the condition that the biased type of the user is private and the feedback strategy is adjustable, the application realizes accurate inference of the service state from the strategic biased feedback, the system loss monotonically decreases to zero with the increase of the number of users, and is suitable for online service evaluation, real-time traffic reporting, network quality monitoring and other crowd-sourcing scenes.
Need to check novelty before this filing date? Find Prior Art