首页 | 本学科首页   官方微博 | 高级检索  
     检索      

一种局部属性加权朴素贝叶斯分类算法
引用本文:张伟,王志海,原继东,刘海洋.一种局部属性加权朴素贝叶斯分类算法[J].北京交通大学学报(自然科学版),2018,42(2):14-21.
作者姓名:张伟  王志海  原继东  刘海洋
作者单位:北京交通大学计算机与信息技术学院,北京,100044;北京交通大学计算机与信息技术学院,北京,100044;北京交通大学计算机与信息技术学院,北京,100044;北京交通大学计算机与信息技术学院,北京,100044
基金项目:国家自然科学基金(61771058;61702030),北京市自然科学基金(4182052),中央高校基本科研业务费专项资金(2017YJS036)National Natural Science Foundation of China(61771058;61702030),Beijing Municipal Natural Science Foundation(4182052),Fundamental Research Funds for the Central Universities(2017YJS036)
摘    要:朴素贝叶斯模型具有的简单性和有效性,使其在诸多问题领域表现出优良的性能,但其属性条件独立性假设在实际应用中难以成立.而属性加权是降低属性条件独立性假设对分类器性能影响的主要途径.传统建立在整个数据集上的单一全局模型忽略了每个测试实例所具有的特点,同时从整个训练集上学习到的属性权重并不能准确反映每个属性对待分类实例的影响.为此提出一种基于数据驱动的懒惰式局部属性加权方法,它在每个测试实例的近邻集合上学习属性权重,并通过最优化方法建立相应的局部属性加权朴素贝叶斯模型.实验结果表明:和当前常见的准朴素贝叶斯模型相比,本文模型具有较高的分类准确率.

关 键 词:朴素贝叶斯  懒惰式  属性加权  局部加权

A locally attribute weighted naive Bayes classifier
ZHANG Wei,WANG Zhihai,YUAN Jidong,LIU Haiyang.A locally attribute weighted naive Bayes classifier[J].JOURNAL OF BEIJING JIAOTONG UNIVERSITY,2018,42(2):14-21.
Authors:ZHANG Wei  WANG Zhihai  YUAN Jidong  LIU Haiyang
Abstract:Naive Bayes(NB) classifier has exhibited excellent performance on many problem domains due to its simplicity and efficiency.In reality the conditional independence assumption of Naive Bayes isn't always true.Attribute weighting is one of the most popular methods to alleviate this assumption's influence on classification results.However,traditional classification models ignore characteristics of each test instance,and the weight vector learned from the whole training set failed to reflect each attribute's contribution of distinguishing each test instance correctly.To this end,a data driven lazy learning locally attribute weighted naive Bayes model is proposed.The attribute weights for each test instance are learned from its neighborhoods,and learned weights are employed to build the locally weighted model by optimization method.Experimental results on benchmark datasets demonstrate that the proposed approach is more accurate than other classical classifiers.
Keywords:
本文献已被 CNKI 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号