Intelligent Crowdsourcing Based on Online Social Networks

The prevalence of Online Social Networks has completely changed how individuals communicate and cooperate. This dissertation explores the integration of intelligent crowdsourcing mechanisms into OSNs, focusing on game theory, reinforcement learn- ing, and social network dynamics to enhance the allocation of tasks, participation by users, and distribution of rewards. It proposes several models that utilize trust- based models, hedonic coalition games, and influencer dynamics for optimization of crowdsourcing. The major challenges to be addressed include task selection, reward distribution, and incentivizing participation, taking into consideration aspects re- lated to social influence, trust, and user engagement. The models so proposed have shown scalability, operational efficiency, and outperform the traditional crowdsourcing methods using simulations and case studies. This work takes up the challenge of integrating OSNs with crowdsourcing and develops solutions to enable users and platforms to cooperatively solve complex tasks in a decentralized and trust-aware manner.