Offline Worker Selection for Real-Time Spatial Crowdsourcing Multi-Worker Tasks

Abstract

Spatial crowdsourcing consists of location-specific tasks that require people to be physically at specific locations to complete them. In this paper we focus on worker selection for spatial crowdsourcing where each task requires multiple workers to accomplish. We mathematically formulate the problem and prove its APX-hardness. We develop efficient greedy algorithms with a good approximation ratio. Compared with state-of-the art approach, our proposed algorithm outperforms by 35%.