用Matlab语言实现BNC  

Constructing BNC using Matlab

作　　者：程泽凯[1] 林士敏[1] 

机构地区：[1]广西师范大学计算机科学系

出　　处：《安徽工业大学学报（自然科学版）》2004年第4期324-327,共4页Journal of Anhui University of Technology（Natural Science）

摘　　要：分类技术是数据采掘的基础与核心,建构分类器是分类技术的关键,利用贝叶斯网络可以构造出分类性能较好的分类器。基于BN Toolkit(BNT)软件包利用Matlab语言实现了两种贝叶斯网络分类器(BNC)。分别基于GS算法和K2算法学习分类器结构。用UCI(University of california in Irvine)上下载的标准数据集验证所建构的BNC,实验结果表明所建构BNC的分类准确率高于文献中所列的NBC和TANC结果,从而表明所建立分类器的有效性和正确性。最后列出了进一步要做的工作。Classification technique has been considered as foundation and a hotspot research in data mining. Constructing classifier plays a key role in classification technique and effective classifier can be constructed using Bayesian Networks.BNC has been constructed using Matlab based BNT software package. There includes two kinds of classifiers that based GS algorithm and K2 algorithm. Effective performance have been achieved in experiments for standard data set from UCI. Experiment results show that a classifier obtained by BNC is superior to NBC and TANC in literature. It demonstrates the effectiveness and correctness of these classifiers. Next works are listed in the end.

关 键 词：贝叶斯网络 贝叶斯网络分类器 MATLAB应用 数据采掘 

分 类 号：TP39[自动化与计算机技术—计算机应用技术]