Research On Task Allocation Algorithm Of Crowdsourcing Platform Based On Game Theory

With the rapid development of information technology,crowdsourcing has emerged as a new business model that integrates group wisdom and solves problems flexibly and effectively,and has promoted the development of crowdsourcing platforms with its unique advantage of "low cost and high return".More and more individuals,companies and organizations are using this model to save costs.However,in a dynamic and complex crowdsourcing platform,there are a large number of complex tasks arising from various complex scenarios that workers cannot complete on their own.The coordination among workers becomes more difficult,the benefits of individual tasks cannot be properly distributed,workers’ motivation to participate decreases with lower pay,and workers’ interests are never aligned with the platform,so a more fair and wellconsidered task allocation mechanism is needed.In this paper,we take the self-interest of workers as the entry point and game theory as the basis of the task allocation of crowdsourcing platform,and focus on the impact of workers’ responses based on selfinterest on their own benefits and platform benefits.Three efficient task allocation methods are designed to solve the problems of the current task allocation algorithm of crowdsourcing platform.（1）To address the problem of unreasonable distribution of individual task benefits in crowdsourcing platforms,an algorithm called Task Allocation Based on Nash Bargaining Solution（TANBS）is proposed.First,the interaction between workers and crowdsourcing platforms is analyzed using the Nash bargaining solution theory in game theory,and the interaction between crowdsourcing platforms and workers is described as a one-to-many bargaining relationship.Then the influence of individual rationality on benefit allocation and task assignment is discussed.Finally,the effectiveness of the algorithm is verified by simulation experiments.（2）To address the problem that low motivation of task participants in crowdsourcing platforms leads to low total platform benefits,a point mechanism is designed to help crowdsourcing platforms improve workers’ motivation to complete tasks through credibility incentives,and the proposed algorithm is called Task Allocation Algorithm Considering Worker Participation（TAACWP）.First,the value orientation of workers in accepting tasks in different task environments is analyzed.Secondly,an incentive of platform points is designed based on the perspective provided by motivation in game mechanics.Finally,the results of the simulation experiments show that the incentive mechanism is effective in increasing user engagement and total platform revenue.（3）To address the problem that inconsistent interests between task participants and the platform in crowdsourcing platforms cannot maximize the total social gain.In this paper,we propose a two-stage Task Allocation Algorithm Based on Market Weights（TSABMW）,which aims to help both the platform and workers maximize profits while minimizing costs,and complete the transformation from a zero-sum game to a cooperative game.First,the optimal response strategy is used to make the worker’s payoff matrix reach Nash equilibrium.Second,the concept of market weights is proposed,the leaders and followers in the coalition are defined,and the rules of worker coalition formation and cooperation in the second stage are constructed.The final simulation results show that the algorithm not only allows task candidates to fully exploit their advantages to compete,but also optimizes the overall market gains.