Task Assignment with Spatio-temporal Recommendation in Spatial Crowdsourcing

With the development of GPS-enabled smart devices and wireless networks, spatial crowdsourcing has received wide attention in assigning location-sensitive tasks to moving workers. In real-world scenarios, workers may show different preferences in different spatio-temporal contexts for the assigned tasks. It is a challenge to meet the spatio-temporal preferences of workers when assigning tasks. To this end, we propose a novel spatio-temporal preference-aware task assignment framework which consists of a translation-based recommendation phase and a task assignment phase. Specifically, in the first phase, we use a translation-based recommendation model to learn spatio-temporal effects from the workers’ historical task-performing activities and then calculate the spatio-temporal preference scores of workers. In the task assignment phase, we design a basic greedy algorithm and a Kuhn-Munkras (KM)-based algorithm which could achieve a better balance to maximize the total rewards and meet the spatio-temporal preferences of workers. Finally, extensive experiments are conducted, verifying the effectiveness and practicality of the proposed solutions.