贝叶斯网络能很好的量化复杂系统中普遍存在的不确定性因素，同时基于贝
叶斯网络的机器学习算法使建立的系统模型具有很好的智能性。

本文在收集大量欧亿·体育（中国）有限公司的基础上对贝叶斯网络的表示、学习和推理进行深入的
研究和探讨，将贝叶斯网络与其他智能方法结合应用于通信网络设备的故障诊断
中。该方法应用到通信网络系统的故障诊断实例中，结果表明该方法在进行系统
可靠性分析时能够充分利用系统的模糊信息和不确定信息，从而提高故障诊断的
效率。

关键词：贝叶斯网络；故障树；故障诊断；管理信息系统黑龙江大学硕士学位论文
-II-
Abstract
Due to the particularity of complexity and running environment of
communication network structure, the failure rate has been relatively high, and the
failure is also a great harm. Therefore, the fault diagnosis communication network
equipment has always been an important aspect of the application of fault diagnosis
technology.
Bayesian network can exist well quantified complex factors of uncertainty in the
system, at the same time Bayesian network based machine learning algorithm to make
the system model of the intelligent well.
In this paper, based on the data collected on the Bayesian network representation,
learning and reasoning of fault diagnosis method, the Bayesian network and other
intelligence applied to communications network equipment. Fault diagnosis examples,
the method is applied to the communication network of the system, the results show
that the method can make full use of the fuzzy information of the system and the
uncertain information in the analysis of system reliability, so as to improve the
efficiency of failure diagnosis.
Keywords: Bayesian networks; fault tree; fault diagnosis; management information
systems