An Evolutionary Algorithm for Collaborative Mobile Crowdsourcing Recruitment in Socially Connected IoT Systems

Mobile crowdsourcing (MCS) enables a distributed problem-solving model in which a crowd of smart devices' users is engaged in the task of solving a data sensing problem through an open call. With the increasing complexity of tasks that are crowdsourced and the need of collaboration among workers, collaborative MCS (CMCS) has emerged to enable requesters to form teams of skilled IoT workers and promote their ability to cooperate together. To efficiently execute such tasks, optimizing the team recruitment process must be conducted. In this paper, we design a low complexity CMCS team recruitment approach that forms and hires a group of socially connected workers having sufficient skills to accomplish a CMCS task. Inspired from swam intelligence, the proposed recruitment approach enables project matching and virtual team formation according to four different fuzzy-logic-based criteria: level of expertise, social relationship strength, recruitment cost, and platform's confidence level. Applied to a real-world data set, experimental results illustrate the performances of the proposed genetic algorithm for CMCS recruitment and show that our approach outperforms the meta-heuristic particle swarm optimization algorithm. Moreover, it is shown that the proposed approach achieves close performance to those of the baseline optimal integer linear program with significant computational saving.