A Task Matching Algorithm Based On Reward Model For Spatial Crowdsourcing

The maturity of wireless network and the popularity of mobile devices provide strong support for implementing spatial crowdsourcing,allowing people on the Internet to complete time-constrained and location-based real tasks in the physical world.There are many simple and trivial tasks in real life that are difficult to implement with simple manual tasks.But everyone has their own limitations and can't see the overall situation better.At this time,we need a "person" who can grasp the overall situation."Person" can assign the appropriate job to the appropriate person according to the overall situation,so that better results can be achieved.Such "Person" have crowdsourcing ideas.The crowdsourcing process that takes time and place information into account is called spatial crowdsourcing.How to assign tasks to the appropriate workers in spatial crowdsourcing is a very core and difficult problem.The goal of it is to assign spatial tasks to more appropriate workers on the basis of space and time saving.Inspired by this situation,this paper studies the problem of space-time task assignment.The main research results are as follows:(1)The research status and related theories of current space-time task assignment problems are summarized.(2)Spatial crowdsourcing task matching algorithm based on compensation model is proposed,and experiments are carried out on the two real-world datasets of Gowalla and YELP through the CPLEX constraint solver.The results show that the proposed method has higher performance in task matching rate and matching rationality,and can provide optimization solutions for the allocation of space-time tasks.(3)Finally,in order to solve the practical problem of over-limit transportation,the Eyeon system using the idea of spatial crowdsourcing is proposed and designed.It aims to mobilize the whole society to participate in the supervision of over-limit transportation.