Research On Task Assignment Strategy Based On Spatial Dynamics In Crowdsourcing

Crowdsourcing stems from the shared economy,and it is paired with over-utilized or underutilized resources and demands.Crowdsourcing-based Logistics integrates idle resources through information technology to create a cost-effective logistics system.As crowdsourcing-based logistics involves networks and mobile terminals,which poses a huge challenge to crowdsourcingbased logistics platform.Different from traditional logistics models,the tasks and resources of crowdsourcing-based logistics have more social attributes.In addition,the crowdsourcing-based logistics platform needs high timeliness,accuracy and reliability.Therefore,in order to meet the requirements of quality and reliability in logistics crowdsourcing,an efficient and reliable task allocation mechanism becomes a core issue of crowdsourcing-based logistics.This paper mainly studies the optimization architecture and task assignment strategy of the spatial crowdsourcing platform,then focuses on the reliability of dynamic crowdsourcing workers,also puts forward an effective task allocation strategy for crowdsourcing logistics platform to ensure the reliable delivery of logistics information data.The main contributions can be summarized as follows:(1)Aiming at the reliability of crowdsourcing workers in crowdsourcing-based logistics platform,a learning approach in crowdsourcing based on ability update is proposed.This chapter introduces the capture factor,which is used to do differential analysis according to the value of capture factor.The proposed approach can set the optimal strategy for always selecting high quality labelers and make these labelers to accomplish the assigned tasks according to the estimated quality.The proposed approach can be widely used to assign tasks in crowdsourcing.Simulation results show that this online learning approach based on ability update can improve quality(accuracy)and efficiency(time)in crowdsourcing.(2)Aiming at the problem of tasks matching with workers in the crowdsourcing-based logistics platform,this chapter first studies a practical problem of task assignment,namely reliability aware spatial crowdsourcing(RA-SC),which takes the constrained tasks and numerous dynamic workers into consideration.Specifically,the worker confidence is introduced to reflect the completion reliability of the assigned task.Our RA-SC problem is to perform task assignments such that the reliability under budget constraints is maximized.Then,we reveal the typical property of the proposed problem,and design an effective strategy to achieve a high reliability of the task assignment.Extensive experimental results also demonstrate that the proposed strategyis stable and effective for spatial crowdsourcing.(3)Aiming at the efficiency of data forwarding and delivery in crowdsourcing-based logistics,this chapter studies how mobile crowdsourcing affects the content-centric network(CCN)and proposes a crowdsourcing-based content-centric network.The target is to take advantage of crowdsourcing to achieve better delivery of logistics content in CCN.Based on characteristics of social attributes(i.e.,social trust and social reciprocity),the enhancement effect of social characteristics in crowdsourcing and the efficient collaboration among devices are fully tapped.Simulation results show that the proposed network architecture can adapt to crowdsourcing-based logistics scenarios and effectively improve the delivery of logistics data.