¹Computer Science Department, Shijiazhuang University, Shijiazhuang, China
²Computer Engineering Department, Ordnance Engineering College, Shijiazhuang, China
³Computer Science Department, Nanjing Normal University, Nanjing, China

---

Abstract—Dynamic Bayesian network is the extension of Bayesian network in solving time series problems .It can be well dealt with the time-varying multivariable problem. A state model is given based on Dynamic Bayesian network. The model can more accurately describe the relationship between the system state and the influencing factors. Single-step and multi-step prediction algorithms are given to predict the system state. The multi-step state prediction algorithm is achieved by extending time-slice. In this paper, the width of the reasoning is used to simplify the amount of data in the reasoning process.

Index Terms—Dynamic Bayesian networks, state prediction model, single-step prediction algorithm, multi-step prediction algorithm.

[PDF]

Cite: Zili Zhang, Hongwei Song, Yan Li, Yan Li, "Prediction Algorithm for State Prediction Model," Journal of Computers vol. 7, no. 2, pp. 507-513, 2012.