Spatial Crowdsourcing Task Assignment Based on the Quality of Workers

With the rapid development of mobile Internet, a variety of spatial crowdsourcing platforms have emerged and been widely applied. Task assignment is the core issue of spatial crowdsourcing. The existing methods of task assignment aim at assigning tasks to workers as much as possible, which lacks the guarantee of the quality of the tasks' answer. In this paper, two kinds of task assignment strategy based on the quality of workers are proposed to ensure the accuracy of the answer submitted by the workers as high as possible. The classical quality control algorithm, Incremental Quality Inference, is used to obtain the quality of workers. Capable worker strategy and maximum worker distance-quality strategy are proposed and compared with nearest work strategy to carry out task assignment based on the quality of workers computed by Incremental Quality Inference. Experimental results with discounted data in the offline shopping mall from crowdsourcing platform demonstrate the effectiveness of our approach.