Research And Implementation Of Crowdsourcing Task Assignment Based On Preference Matching

In recent years,the rapid development of Internet technology and open innovation has promoted the popularity of crowdsourcing.Crowdsourcing is an open problemsolving mechanism for the Internet community that aggregates the wisdom of the crowd to better solve problems.Under the crowdsourcing scenario,the task and the receiving workers have different needs and intentions.Assigning tasks to the receiving workers who do not match their needs will affect the completion quality of the crowdsourcing tasks.At the same time,in the process of interaction between the crowd-sourced outsourcer and the receiving worker,the allocation result may change with the discovery of better options by both parties,and the invalid allocation may affect the stability of the allocation result.Therefore,how to effectively match the crowd-sourced tasks with the workers to ensure the effectiveness and stability of the distribution results has become an important issue to ensure the quality of crowdsourcing tasks.However,the existing allocation methods do not consider the bilateral preferences of crowdsourcing users under the condition of stability,and the accuracy of the allocation results needs to be improved.Moreover,there exists the phenomenon that the crowdsourcing task completion quality is low due to the fact that the crowdsourcing contractor or the receiving worker is not satisfied with the current allocation object.In order to solve the above problem,this paper proposes a crowdsourcing task allocation method based on preference matching,this method uses preferences under the condition of stable matching to find satisfaction maximize allocation,reduce invalid number distribution,improve the accuracy and stability of the allocation results,so as to guarantee the completion quality.The method first considers the bidirectional preferences of the crowd-sourcing task and the worker,calculates the satisfaction of the task and the worker according to the order of preference,and generates the satisfaction matrix.Secondly,in order to reduce the number of invalid distributions in the allocation results,the idea of stable matching is used to make both parties of the allocation as satisfied as possible with the current allocation objects on the basis of considering the bilateral preferences of crowdsourcing users,so as to ensure the stability of the allocation results.Then the crowd-sourcing task assignment problem is modeled as an optimization problem to find the maximum satisfaction of tasks under stable matching rules.Finally,greedy algorithm is used to solve the problem and a crowd-sourcing task assignment scheme is obtained.In this paper,the effectiveness of the method is verified by experiments,which show that the method improves the accuracy of allocation scheme and effectively reduces the number of invalid allocations,thus improving the quality of crowdsourcing task completion.In addition,the concept of relational degree is introduced and calculated when dealing with the undifferentiated preferences of the tasks and the workers,so as to reduce the influence of undifferentiated preference on the solving accuracy of the greedy algorithm.The relational degree characteristics in the crowdsourcing environment are verified by experiments.Finally,based on the crowdsourcing task assignment method proposed in this paper,a prototype system of crowd-sourcing task assignment oriented to the satisfaction maximization of preference matching is designed and implemented.