Task Allocation And Incentive Mechanism Design Of Mobile Crowdsourcing With Limited Budget

Mobile crowdsourcing,which uses Internet users to complete some perceptual tasks,is a new model for data collection,analysis and sharing.Task allocation is the core issue of mobile crowdsourcing application research.With a limited budget,it is important to allocate the budget reasonably and match the workers and tasks to achieve the desired goals.In order to encourage users to actively and honestly participate in crowdsourcing activities and stay active in the crowdsourcing system for a long time,incentive mechanism is a key to the development of mobile crowdsourcing.In the existing research on mobile crowdsourcing task allocation considering budget,most studies assume that the quality level of workers for all tasks is single,ignoring the diversity of workers' abilities,and at the same time not considering the real cost of workers,which leads to the low overall quality of task completion and the waste of budget.Moreover,most of the existing mobile crowdsourcing tasks are assigned at one time,lacking the negotiation process between workers and task initiators,which leads to too rigid task requirements and fails to maximize the overall social welfare of workers and task initiators.In view of the above problems,this paper studies the task allocation and incentive mechanism design of mobile crowdsourcing with limited budget,and the specific work is as follows:(1)This paper proposes a task allocation and incentive mechanism based on task classification,and establishes a task difficulty model and a worker quality model.With the goal of reducing the task difficulty,and encouraging workers to quote at real cost,the utility of task initiators can be maximized within a limited budget.Through theoretical analysis and comparative experiments,it is verified that the mechanism proposed in this paper is truthful,individual rational and budget feasible,and it effectively improves the task perception quality of workers.(2)This paper proposes a task allocation and incentive mechanism based on multi-attribute concession negotiation.This mechanism increases the interaction between task initiators and workers by establishing a multi-round bargaining model for the negotiable attributes of tasks,and solves the problem that the task demand is too rigid due to one-time task allocation.Then,the concept of reputation value is introduced to motivate workers to honestly complete tasks according to the negotiation results.It is proved by theory and experiment that this mechanism can maximize social welfare,and has the characteristics of individual rationality,incentive compatibility and budget feasible.(3)Combining with the actual situation,this paper analyzes the components of the mobile crowdsourcing platform,and develops a mobile crowdsourcing system FPFW(find pros and find works)for task publishing and answer collection with Python language.The tasks are divided into two types:hard tasks and negotiable tasks,which correspond to the two task allocation mechanisms proposed in points(1)and(2),respectively,to realize the matching between tasks and workers.