基于泊松分布的加权朴素贝叶斯文本分类算法

doi:10.19678/j.issn.1000-3428.0054056

计算机工程 ›› 2020, Vol. 46 ›› Issue (4): 91-96. doi: 10.19678/j.issn.1000-3428.0054056

基于泊松分布的加权朴素贝叶斯文本分类算法

赵博文, 王灵矫, 郭华

湘潭大学信息工程学院, 湖南湘潭 411105

收稿日期:2019-03-01 修回日期:2019-04-18 出版日期:2020-04-15 发布日期:2019-05-27
作者简介:赵博文(1994-),男,硕士研究生,主研方向为数据挖掘;王灵矫,副教授、博士;郭华,高级实验师、硕士。
基金资助:
国家自然科学基金（61771414）。

Weighted Naive Bayes Text Classification Algorithm Based on Poisson Distribution

ZHAO Bowen, WANG Lingjiao, GUO Hua

College of Information Engineering, Xiangtan University, Xiangtan, Hunan 411105, China

Received:2019-03-01 Revised:2019-04-18 Online:2020-04-15 Published:2019-05-27

摘要/Abstract

摘要： 朴素贝叶斯（NB）算法应用于文本分类时具有简单性和高效性，但算法中属性独立性与重要性一致的假设，使其在精确度方面存在瓶颈。针对该问题，提出一种基于泊松分布的特征加权NB文本分类算法。结合泊松分布模型和NB算法，将泊松随机变量引入特征词权重，在此基础上定义信息增益率对文本特征词加权，削弱传统算法属性独立性假设造成的影响。在20-newsgroups数据集上的实验结果表明，与传统NB算法及其改进算法R_W，C-MNB和CFSNB相比，该算法可使文本分类的准确率、召回率和F₁值得到提升，并且执行效率高于K-最近邻算法和支持向量机算法。

关键词: 文本分类, 朴素贝叶斯算法, 泊松分布, 信息增益率, 特征词权重

Abstract: Naive Bayes(NB) algorithm is simple and efficient when applied to text classification,but it has a bottleneck in accuracy due to the intrinsic assumption that attribute independence and attribute importance are consistent.To solve this problem,this paper proposes a feature-weighted NB text classification algorithm based on Poisson distribution.The algorithm combines the Poisson distribution model with the NB algorithm,and the Poisson random variable is introduced into the weight of feature words.On this basis,the Information Gain Ratio(IGR) is defined to weigh the feature words of texts,and thus the effects of the attribute independence assumption of traditional algorithms can be reduced.Experimental results on the 20-newsgroups data set show that,compared with NB algorithm and its improved algorithms R_W,C-MNB and CFSNB,this algorithm can improve the accuracy rate,recall rate and F₁ value of text classification.Meanwhile,its execution efficiency is higher than K-Nearest Neighbor(KNN) algorithm and Support Vector Machine(SVM) algorithm.

Key words: text classification, Naive Bayes(NB) algorithm, Poisson distribution, Information Gain Rate(IGR), weight of feature words

中图分类号:

TP391

赵博文, 王灵矫, 郭华. 基于泊松分布的加权朴素贝叶斯文本分类算法[J]. 计算机工程, 2020, 46(4): 91-96.

ZHAO Bowen, WANG Lingjiao, GUO Hua. Weighted Naive Bayes Text Classification Algorithm Based on Poisson Distribution[J]. Computer Engineering, 2020, 46(4): 91-96.

https://www.ecice06.com/CN/Y2020/V46/I4/91

图/表 4

参考文献 20

[1]	LEE C H.A gradient approach for value weighted classification learning in naive Bayes[J].Knowledge-Based Systems,2015,85(1):71-79.
[2]	XIANG Z L,YU X R,KANG D K.Bayesian prediction model based on attribute weighting and kernel density estimations[J].Mathematical Problems in Engineering,2015(7):1-7.
[3]	AGRAWAL R,IMIELINSKI T,SWAMI A.Database mining:a performance perspective[J].IEEE Transactions on Knowledge and Data Engineering,1993,5(6):914-925.
[4]	FRIEDMAN N,GEIGER D,GOLDSZMIDT M.Bayesian network classifiers[J].Machine Learning,1997,29(2):131-163.
[5]	LIANG Cong,XIA Shuyin,CHEN Zizhong.Improvement k-nearest neighbor classification algorithm based on reference points[J].Computer Engineering,2019,45(2):167-172.(in Chinese)梁聪,夏书银,陈子忠.基于参考点的改进k近邻分类算法[J].计算机工程,2019,45(2):167-172.
[6]	CUI Jianming,LIU Jianming,LIAO Zhouyu.Research of text categorization based on support vector machine[J].Computer Simulation,2013,30(2):299-302.(in Chinese)崔建明,刘建明,廖周宇.基于SVM算法的文本分类技术研究[J].计算机仿真,2013,30(2):299-302.
[7]	RAHMAN A,QAMAR U.A Bayesian classifiers based combination model for automatic text classification[C]//Proceedings of IEEE International Conference on Software Engineering and Service Science.Washington D.C.,USA:IEEE Press,2017:63-67.
[8]	UYSAL A K.An improved global feature selection scheme for text classification[J].Expert Systems with Applications,2016,43(C):82-92.
[9]	WANG Shasha,JIANG Liangxiao,LI Chaoqun.Adapting naive Bayes tree for text classification[J].Knowledge and Information Systems,2015,44(1):77-89.
[10]	JIANG Liangxiao,LI Chaoqun,WANG Shasha,et al.Deep feature weighting for naive Bayes and its application to text classification[J].Engineering Applications of Artificial Intelligence,2016,52(C):26-39.
[11]	ZHANG Chunying,WANG Jing.A new weighted naïve Bayesian classification algorithm[J].Microcomputer Information,2010,26(10):222-224.
[12]	LANGLEY P,SAGE S.Induction of selective Bayesian classifiers[C]//Proceedings of the 10th International Conference on Uncertainty in Artificial Intelligence.Seattle,USA:[s.n.],1994:399-406.
[13]	QI Yaun,QIAO Yu.Text categorization based on naive Bayesian[J].Electronic Science and Technology,2017,4(5):114-116.(in Chinese)漆原,乔宇.针对朴素贝叶斯文本分类方法的改进[J].电子科学技术,2017,4(5):114-116.
[14]	WU Jianjun,LI Changbing.Mutual information-based weighted naive Bayes text classification algorithm[J].Computer Systems and Applications,2017,26(7):178-182.(in Chinese)武建军,李昌兵.基于互信息的加权朴素贝叶斯文本分类算法[J].计算机系统应用,2017,26(7):178-182.
[15]	BHARATHI V,JAYANIVETHA P K,KANNIGA K,et al.A Bayesian classification approach using class-specific features for text categorization[J].International Journal of Modern Trends in Engineering and Science,2017,4(3):103-107.
[16]	LI Y,LUO C,CHUNG S M.Weighted naive Bayes for text classification using positive term-class dependency[J].International Journal on Artificial Intelligence Tools,2012,21(1):1-8.
[17]	WANG Shasha,JIANG Liangxiao,LI Chaoqun.A CFS-based feature weighting approach to naïve Bayes text classifiers[C]//Proceedings of ICANN'14.Berlin,Germany:Springer,2014:555-562.
[18]	ZHANG Huiming,LI Bo.Characterizations of discrete compound Poisson distributions[J].Communications in Statistics-Theory and Methods,2016,45(22):6789-6802.
[19]	AHMAD M.A short note on Conway-Maxwell-Hyper Poisson distribution[J].Journal of the Royal Statistical Society Series C-Applied Statistics,2007,54(1):127-142.
[20]	LI Kaiqi,DIAO Xingchun,CAO Jianjun.Improved algorithm of text feature weighting based on information gain[J].Computer Engineering,2011,37(1):16-18.(in Chinese)李凯齐,刁兴春,曹建军.基于信息增益的文本特征权重改进算法[J].计算机工程,2011,37(1):16-18.

选择文件类型/文献管理软件名称

选择包含的内容

基于泊松分布的加权朴素贝叶斯文本分类算法

Weighted Naive Bayes Text Classification Algorithm Based on Poisson Distribution

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 4

参考文献 20

相关文章 15

编辑推荐

Metrics

本文评价

[1]	钱来, 赵卫伟. 基于对比学习和注意力机制的文本分类方法[J]. 计算机工程, 2024, 50(7): 104-111.
[2]	游奔, 李晓红, 姚锦, 冯绍杰. 基于多粒度图与注意力机制的半监督短文本分类[J]. 计算机工程, 2024, 50(5): 83-90.
[3]	张博旭, 蒲智, 程曦. 基于提示学习的维吾尔语文本分类研究[J]. 计算机工程, 2023, 49(6): 292-299,313.
[4]	王春东, 孙嘉琪, 杨文军. 基于矫正理解的中文文本对抗样本生成方法[J]. 计算机工程, 2023, 49(2): 37-45.
[5]	陈可嘉, 刘惠. 基于改进BiGRU-CNN的中文文本分类方法[J]. 计算机工程, 2022, 48(5): 59-66,73.
[6]	李冉冉, 刘大明, 刘正, 常高祥. 融合笔画特征的胶囊网络文本分类[J]. 计算机工程, 2022, 48(3): 69-73,80.
[7]	武娇, 洪彩凤, 顾永春, 顾兴全, 金世举. 基于类邻域字典的线性回归文本分类[J]. 计算机工程, 2021, 47(8): 93-99,108.
[8]	周伟枭, 蓝雯飞. 融合文本分类的多任务学习摘要模型[J]. 计算机工程, 2021, 47(4): 48-55.
[9]	彭俊利, 谷雨, 张震, 耿小航. 融合单词贡献度与Word2Vec词向量的文档表示[J]. 计算机工程, 2021, 47(4): 62-67.
[10]	何力, 郑灶贤, 项凤涛, 吴建宅, 谭林. 基于深度学习的文本分类技术研究进展[J]. 计算机工程, 2021, 47(2): 1-11.
[11]	袁自勇, 高曙, 曹姣, 陈良臣. 基于异构图卷积网络的小样本短文本分类方法[J]. 计算机工程, 2021, 47(12): 87-94.
[12]	段丹丹, 唐加山, 温勇, 袁克海. 基于BERT模型的中文短文本分类算法[J]. 计算机工程, 2021, 47(1): 79-86.
[13]	丁辰晖, 夏鸿斌, 刘渊. 融合知识图谱与注意力机制的短文本分类模型[J]. 计算机工程, 2021, 47(1): 94-100.
[14]	方瑞, 于俊洋, 董李锋. 基于特征矩阵构造与BP神经网络的垃圾文本过滤模型[J]. 计算机工程, 2020, 46(8): 271-276.
[15]	王芝辉, 王晓东. 基于神经网络的文本分类方法研究[J]. 计算机工程, 2020, 46(3): 11-17.

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

基于泊松分布的加权朴素贝叶斯文本分类算法

Weighted Naive Bayes Text Classification Algorithm Based on Poisson Distribution

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 4

参考文献 20

相关文章 15

编辑推荐

Metrics

本文评价