A Deep Reinforcement Learning Approach for Composing Moving IoT Services

We develop a novel framework for efficiently and effectively discovering crowdsourced services that move in close proximity to user over period of time. introduce moving service model which is modelled as region. propose deep reinforcement learning-based composition approach select compose IoT considering quality parameters. Additionally, we parallel flock-based discovery algorithm ground-truth measure the accuracy proposed approach. The experiments on two real-world datasets verify effectiveness efficiency