Mobile Crowdsensing

Mobile crowdsensing leverages the collective sensing capabilities of mobile devices to gather large-scale data for various applications, primarily aiming to improve data collection efficiency and reduce costs compared to traditional methods. Current research focuses on addressing challenges like data sparsity and quality through advanced algorithms such as deep matrix factorization, recurrent neural networks, and generative AI models, often incorporating techniques from multi-armed bandits and federated learning to optimize resource allocation and incentivize participation. This approach holds significant potential for improving urban management (e.g., traffic flow estimation, parking management), environmental monitoring, and disaster response, while also advancing the fields of machine learning and distributed systems.