Internet of things(Io T) is an emerging technique that offers advanced connectivity of devices, systems, services, and human beings. With the rapid development of hardware and network technologies, the Io T can refer to a wide variety and large number of devices, resulting in complex relationships among Io T devices. The dependencies among Io T devices, which reflect their relationships, are with reference value for the design, development and management of Io T devices. This paper proposes a stochastic model based approach for evaluating the dependencies of Io T devices. A random walk model is proposed to describe the relationships of Io T devices, and its corresponding Markov chain is obtained for dependency analysis. A framework as well as schemes and algorithms for dependency evaluation in real-world Io T are designed based on traffic measurement. Simulation experiments based on real-life data extracted from smart home environments are conducted to illustrate the efficacy of the approach. Internet of things (Io T) is an emerging technique that offers advanced connectivity of devices, systems, services, and human beings. With the rapid development of hardware and network technologies, the Io T can refer to a wide variety and large number of devices , resulting in complex relationships among Io T devices. which dependencies among Io T devices, which reflect their relationships, are with reference value for the design, development and management of Io T devices. This paper proposes a stochastic model based approach for evaluating the dependencies of Io T devices. A random walk model is proposed to describe the relationships of Io T devices, and its corresponding Markov chain is obtained for dependency analysis. A framework as well as schemes and algorithms for dependency evaluation in real-world Io T are designed based on traffic measurement. Simulation experiments based on real-life data extracted from smart home environments are to illustrate the efficacy of the ap proach.