基于三对角矩阵的完全贝叶斯分类器研究

针对连续属性朴素贝叶斯分类器不能有效利用属性之间的条件依赖信息、而对其进行依赖扩展中的高阶协方差矩阵的求逆和行列式运算又非常困难等问题,将三对角矩阵和多元高斯函数相结合,建立连续属性完全贝叶斯分类器,并在三对角矩阵中引入平滑参数,通过对平滑参数的调整来实现分类器的优化。使用UCI数据的实验结果显示,经过优化的连续属性完全贝叶斯分类器具有良好的分类准确性。...

Full description

Saved in:
Bibliographic Details
Published in计算机应用研究 Vol. 32; no. 3; pp. 740 - 742
Main Author 冷翠平 王双成 杜瑞杰
Format Journal Article
LanguageChinese
Published 上海立信会计学院开放经济与贸易研究中心,上海201620%上海立信会计学院开放经济与贸易研究中心,上海,201620 2015
上海立信会计学院数学与信息学院,上海,201620%上海立信会计学院数学与信息学院,上海201620
Subjects
三对角矩阵
多元高斯函数
完全贝叶斯分类器
平滑参数
朴素贝叶斯分类器
full Bayesian classifier
multivariate Gaussian function
三对角矩阵
平滑参数
完全贝叶斯分类器
tridiagonal matrix
naive Bayes classifiers
smoothing parameters
朴素贝叶斯分类器
多元高斯函数
Online AccessGet full text

Cover

More Information
Summary:针对连续属性朴素贝叶斯分类器不能有效利用属性之间的条件依赖信息、而对其进行依赖扩展中的高阶协方差矩阵的求逆和行列式运算又非常困难等问题,将三对角矩阵和多元高斯函数相结合,建立连续属性完全贝叶斯分类器,并在三对角矩阵中引入平滑参数,通过对平滑参数的调整来实现分类器的优化。使用UCI数据的实验结果显示,经过优化的连续属性完全贝叶斯分类器具有良好的分类准确性。
Bibliography:51-1196/TP
LENG Cui-ping, WANG Shuang-chdng, DU Rui-jie （ a. School of Mathematics ＆ Information, b. Open Economic ＆ Trade Research Center, Shanghai Lixin University of Commerce, Shanghai 201620, China）
naive Bayes classifiers ; full Bayesian classifier ; multivariate Gaussian function; tridiagonal matrix; smoothingparameters
Naive Bayes classifiers with continuous attributes can not make full used of dependency information between attrib- utes. And in their dependency extension, it is difficult to calculate the inverse and the determinant of the high-order covari- ance matrix. To solve the dilemma, combining the tridiagonal matrix and multivariate Gaussian function can help build the full Bayesian classifiers with continuous attributesl And introduced smoothing parameters to tridiagonal matrix, which can contrib- ute to the classifiers optimization. Experimental results resulting from using UCI data show that the optimized classifiers with continuous attributes have good classification accuracy.
ISSN:1001-3695
DOI:10.3969/j.issn.1001-3695.2015.03.023