《基于Matlab的贝叶斯网络学习方法研究》-毕业论文.docVIP

PAGE 精品 精品 齐 齐 哈 尔 大 学 毕业设计（论文） 题 目 基于Matlab的贝叶斯网络学习方法研究__ 学 院 计算机与控制工程学院_________________ 专业班级 计本 ___________________________ 学生姓名 _______________________________ 指导教师 _______________________________ 成 绩 摘 要 贝叶斯网络提供了一种表达因果关系的方法。它结合图模型理论和统计学来表达随机变量之间的不确定性知识，成为目前不确定知识表达和推理领域最有效的模型之一。贝叶斯网络学习主要包括：结构学习和参数学习，通过网络结构与数据集可以确定参数，因此结构学习是贝叶斯网络学习的核心，有效的结构学习方法和算法是构建最优网络结构的基础。 本文在对贝叶斯网络的起源、发展、特点及应用情况进行介绍的基础上，着重对贝叶斯网络的学习算法进行了研究，阐述了几种贝叶斯网络学习算法的思想。采用MATLAB语言编制的贝叶斯网络工具箱(Bayesian Networks Toolbox, BNT)实现这几种学习算法的应用，并结合实际给出了两个实例进行了说明分析，本文的研究对于充分运用MATLAB的贝叶斯网络工具箱的强大功能,学习和利用贝叶斯方法来解决实际问题有着重要而深远的意义。 关键词：贝叶斯网络；结构学习；参数学习；贝叶斯网络工具箱 Abstract Bayesian network (BN) is an important method for presenting causality and uncertainty among random variables based on graphical model theory and statistics. It is one of the most efficient models in the fields of uncertain knowledge expression and inference. The learning Bayesian networks mainly includes: structural and parameter learning, parameters can be fixed through networks structure and data sets, so structural learning is the core of learning Bayesian network. The efficient structural learning is the basis constructing the most efficient network structure. The paper illustrates the origin, development, characteristics and application of Bayesian networks, emphatically studying Bayesian networks learning algorithm and introducing the theory of some Bayesian networks learning algorithms. The paper implements the application of these learning algorithms with Bayesian Networks Toolbox (BNT) based on MATLAB, and presents two instances to explain and analysis.The research makes full use of the powerful function of BNT based on MATLAB.Learning and using Bayesian method to solve actual problem have important and far-reaching significance. Key words: Bayesian network; Structural learning; Parameter learning; Bayesian Networks Toolbox 目 录 TOC \o 1-3 \h \z \u HYPERLINK \l _