Avoiding Information Leakage in the Formation of Crowdsourcing Teams via Extended Group Multirole Assignment Considering Fairness
Hongze Guo, Haibin Zhu, Dongning Liu · 2024
The development of the Internet has led to the rapid development of a new business model called crowdsourcing. However, increasingly complex crowdsourcing tasks are difficult to be decomposed and decoupled by the performers in the actual execution. The crowdsourcing platform needs to provide a detailed task assignment method to solve this problem. At the same time, crowdsourcing tasks may involve user privacy, and protecting private information from being known by others also needs to be considered by the platform. While completing the crowdsourcing task, considering task fairness and team fairness to improve the reliability of task completion and the fairness perception of crowdsourcing members. Therefore, this article formalizes the crowdsourcing team assignment problem through Environments – Classes, Agents, Roles, Groups, Objects (E-CARGO) model. Introducing information security constraints to construct a new sub-model (GRAINS) to solve the crowdsourcing team assignment problem. On the premise of obtaining the optimal performance assignment, the two types of fairness are discussed, providing a new decision-making scheme for the crowdsourcing platform. Through large-scale random simulation experiments, it is proved that the model can improve task and team fairness while ensuring the overall performance of the task, and quantitatively analyze the partial performance for fairness sacrificed.