Coalition-based task assignment with priority-aware fairness in spatial crowdsourcing

Abstract With the widespread use of networked and geo-positioned mobile devices, e.g., smartphones, Spatial Crowdsourcing (SC), which refers to assignment location-based tasks moving workers, is drawing increasing attention. One critical issues in SC task that allocates appropriate workers. We propose study a novel problem, namely Coalition-based Task Assignment (CTA), where spatial (e.g., home improvement furniture installation) may require more than one worker (forming coalition) cooperate maximize overall rewards design greedy an equilibrium-based CTA approach. The approach forms set coalitions greedily for performing uses acceptance probability identify high-value assignments. In approach, workers form sequence update their strategies (i.e., selecting best-response task), own utility reward coalition they belong to) until Nash equilibrium reached. Since obtained not unique optimal terms total rewards, we further simulated annealing scheme find better equilibrium. To achieve fair assignments, optimize framework distribute fairly among based on marginal contributions give who arrive first at platform highest priority. Extensive experiments demonstrate efficiency effectiveness proposed methods real synthetic data.