图书情报工作

图书情报工作 >

2014 , Vol. 58 >Issue 19: 124 - 128

DOI: https://doi.org/10.13266/j.issn.0252-3116.2014.19.019

知识组织

基于语义网络社团划分的中文文本分类研究

  • 尹丽英 ,
  • 赵捧未
展开
  • 1. 西安电子科技大学经济与管理学院;
    2. 西安邮电大学经济与管理学院
尹丽英，西安电子科技大学经济与管理学院博士研究生，西安邮电大学经济与管理学院讲师，E-mail：yin-liying@163.com；赵捧未，西安电子科技大学经济与管理学院教授，博士生导师。

收稿日期: 2014-06-09

  修回日期: 2014-08-11

  网络出版日期: 2014-10-05

基金资助

本文系国家自然科学基金项目“基于知识地图的对等网语义社区及其知识共享研究”（项目编号：71103138）和中央高校基础科研业务费资助项目“大数据背景下基于用户生成内容的商务智能模型研究”（项目编号：BDY231414）研究成果之一。

收起

A Chinese Text Classification Algorithm Based on Partitioning Community in Semantic Network

  • Yin Liying ,
  • Zhao Pengwei
Expand
  • 1. School of Economics & Management, Xidian University, Xi'an 710071;
    2. College of Economics and Management, Xi'an University of Post & Telecommunications, Xi'an 710121

Received date: 2014-06-09

  Revised date: 2014-08-11

  Online published: 2014-10-05

Fold

摘要

为减少一词多义现象及训练样本的类偏斜问题对分类性能的影响，提出一种基于语义网络社团划分的中文文本分类算法。通过维基百科知识库对文本特征词进行消歧，构建出训练语义复杂网络以表示文本间的语义关系，再次结合节点特性采用K-means算法对训练集进行社团划分以改善类偏斜问题，进而查找待分类文本的最相近社团并以此为基础进行文本分类。实验结果表明，本文所提出的中文文本分类算法是可行的，且具有较好的分类效果。

关键词： 语义网络; 词义消歧; 社团结构; 文本分类

本文引用格式

尹丽英 , 赵捧未 . 基于语义网络社团划分的中文文本分类研究[J]. 图书情报工作, 2014 , 58(19) : 124 -128 . DOI: 10.13266/j.issn.0252-3116.2014.19.019

Abstract

In order to reduce the polysemy phenomenon and the influence of the category deflect problem of training samples, a Chinese text categorization method was proposed on community division of semantic network. Firstly, disambigurtion was in progress through Wikipedia knowledge base, the complex network of text is built in order to represent the semantic relations between training texts. Then, in order to improve the problem of category deflect, the training samples is partitioned by the method of K-means which combined with the synthetic characteristics of network nodes. Finally, the text classification based on the nearest community of testing text is found out according to the nearest community. Results of experiments show that the algorithm proposed by this paper is feasible and can improve the effect of its classification.

Key words： semantic network; word sense disambiguation; community structure; text classification

参考文献

[1] Sebastiani F. Machine learning in automated text categorization[J]. ACM Computing Surveys, 2002, 34(1): 1-47.

[2] 王煜, 白石, 王正欧. 基于特征权重优化的改进 KNN Web 文本分类算法[J]. 情报学报, 2007, 26(5): 643-647.

[3] Yang Y, Liu X. A re-examination of text categorization methods[C]//Proceedings of the 22nd Annual International ACM SIGIR Conference on Research and Development in Information Retrieval. New York:ACM, 1999: 42-49.

[4] Jing Yongxia, Gou Heping, Zhu Yaling. An improved density-based method for reducing training data in KNN[C]//The 2013 International Conference on Computational and Information Sciences. Piscataway:IEEE, 2013: 972-975.

[5] 刘海峰, 姚泽清, 苏展, 等. 文本分类中基于 K-means 的类偏斜 KNN 样本剪裁[J]. 微电子学与计算机, 2012, 29(5): 24-28.

[6] 王超学, 潘正茂, 马春森, 等. 改进型加权 KNN 算法的不平衡数据集分类[J]. 计算机工程, 2012, 38(20): 160-163.

[7] Wei G Y, Zou L, Pan J. Improved text classification algorithm for spam filtering based on CABSOFV[J]. WIT Transactions on Engineering Sciences, 2014, 86:1131-1139.

[8] Liu Zuoguo, Chen Xiaorong. A graph-based text similarity algorithm[C]//2012 National Conference on Information Technology and Computer Science. Beijing:Atlantis Press, 2012.

[9] Giannakopoulos G, Mavridi P, Paliouras G, et al. Representation models for text classification: A comparative analysis over three Web document types[C]//Proceedings of the 2nd International Conference on Web Intelligence, Mining and Semantics. New York:ACM, 2012: 13.

[10] W Jiangning, L Qiaofeng. Research on text similarity computing based on max-common subgraphs[J]. Journal of the China Society for Scientific and Technical Information, 2010, 29(5): 785-791.

[11] 涂新辉, 张红春, 周琨峰,等. 中文维基百科的结构化信息抽取及词语相关度计算方法[J]. 中文信息学报, 2012, 26(3): 109-115.

[12] 赵辉, 刘怀亮, 范云杰. 复杂网络理论在中文文本特征选择中的应用研究[J]. 现代图书情报技术, 2012(9): 23-28.

Options
文章导航

/

〈 〉