A Framework for Optimal Worker Selection in Spatial Crowdsourcing Using Bayesian Network

Spatial Crowdsourcing (SC) is a new paradigm of crowdsourcing applications. Unlike tra-ditional crowdsourcing, SC outsources tasks to distributed potential workers, and those who accept thetask are required to travel to a predefined location to complete it. Currently, the primary aim of SC is tomaximize the number of matched tasks or to minimize the travelling distances of the workers. However,less focus is given in matching the right tasks to the right workers, particularly in a heterogeneous tasksenvironment. To address this lacking, our work provides an efficient framework for selecting optimal workersfor every task with various specification (geographical proximity, domain types, and expiration times), basedon workers’ attributes (task domain-specific knowledge, expertise or performance history, distance to tasklocation, and task workload distribution). We introduce the use of Bayesian Network in modelling andselecting optimal workers, and use k-medoids partitioning technique for tasks clustering and scheduling. Ourexperimental results on both synthetic and real-world large datasets have shown that our proposed approachhas outperformed other baseline approaches, in terms of low average error rate and fast execution time