Multitask Assignment Algorithm Based on Decision Tree in Spatial Crowdsourcing Environment

To improve the resource utilization rate of a platform and increase worker profit, addressing the problem of a limited suitable range in a single-task assignment in a spatial crowdsourcing environment, this paper provides a single-worker multitask assignment strategy. A candidate worker-selection algorithm based on location entropy minimum priority is proposed. Candidate tasks are selected by calculating their location entropy within a selected area. A candidate worker is obtained based on the Manhattan distance between the candidate task and the worker, completing the single-task assignment to the single worker. Then a multitask assignment algorithm based on a decision tree is designed, which builds a multitask screening decision tree and calculates the candidate tasks’ time difference, travel cost ratio, coincidence rate of route, and income growth rate of workers. We filter out the most appropriate task and assign it to a worker to complete the multitasking assignment. Experimental results show that the proposed algorithm can effectively reduce the average travel cost, reduce the idle rate of workers, and improve their income, which has better effectiveness and feasibility.