《计算机应用研究》|Application Research of Computers

不平衡数据集下特征词两面性的新型降维算法

Novel feature selection method based on two-side considering imbalance problem

免费全文下载 (已被下载 次)  
获取PDF全文
作者 付鑫,王洪国,邵增珍,杜秋霞
机构 山东师范大学 信息科学与工程学院 山东省物流优化与预测工程技术研究中心,济南 250014
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2018)07-1947-03
DOI 10.3969/j.issn.1001-3695.2018.07.005
摘要 传统DFS特征选择算法在降维处理时既未考虑样本分布不均的情况,又未涉及负特征词对类别的影响。综合考虑DFS的缺陷并进行优化处理,将DFS与卡方检测算法CHI结合,提出一种改进型特征选择算法DFS-sCHI。引入负特证词作为类别划分的影响因子之一,解决不平衡数据集下所提特征词类别分布不均的问题。经实验分析,不平衡数据集下,DFS-sCHI相比较于DFS在分类精度上有明显提高。
关键词 不平衡数据集;文本分类;特征选择;DFS-sCHI
基金项目 山东省科技发展计划资助项目(2014GGH201022)
山东省经信委软科学计划资助项目(2015EI010)
国家自然科学基金资助项目(71461027)
本文URL http://www.arocmag.com/article/01-2018-07-005.html
英文标题 Novel feature selection method based on two-side considering imbalance problem
作者英文名 Fu Xin, Wang Hongguo, Shao Zengzhen, Du Qiuxia
机构英文名 ShandongProvincialLogisticsOptimization&PredictiveEngineeringTechnologyResearchCenter,SchoolofInformationScience&Engineering,ShandongNormalUniversity,Jinan250014,China
英文摘要 DFS(distinguishing feature selector) neither considered the situation of uneven distribution of samples nor involved the impact of negative words for category. This paper considered the defects of DFS and optimized these defects, combined DFS with CHI to proposed DFS-sCHI that was a improve feature selection method. This method introduced the power of negative words in order to solve the problem that the feature of category was uneven selection. The experimental results indicate that the proposed method improves the classification accuracy obviously.
英文关键词 imbalance data set; text classification; feature selection; DFS-sCHI
参考文献 查看稿件参考文献
  [1] Salton G, Wong A, Yang C S. A vector space model for automatic indexing[J] . Communications of the ACM, 1975, 18(11):613-620.
[2] Yang Yiming, Pedersen J O. A comparative study on feature selection in text categorization[C] //Proc of the 14th International Conference on Machine Learning. San Francisco:Morgan Kaufmann Publishers Inc, 1997:412-420.
[3] Lee C, Lee G G. Information gain and divergence-based feature selection for machine learning-based text categorization[J] . Information Processing & Management, 2006, 42(1):155-165.
[4] Chen Y T, Chen Mengchang. Using Chi-square statistics to measure similarities for text categorization[J] . Expert Systems with Applications, 2011, 38(4):3085-3090.
[5] Liu Huawen, Sun Jigui, Liu Lei, et al. Feature selection with dynamic mutual information[J] . Pattern Recognition, 2009, 42(7):1330-1339.
[6] Shang Wenqian, Huang Houkuan, Zhu Haibin, et al. A novel feature selection algorithm for text categorization[J] . Expert Systems with Applications, 2007, 33(1):1-5.
[7] Ogura H, Amano H, Kondo M. Feature selection with a measure of deviations from Poisson in text categorization[J] . Expert Systems with Applications, 2009, 36(3):6826-6832.
[8] Yang Jieming, Qu Zhaoyang, Liu Zhiying. Improved feature-selection method considering the imbalance problem in text categorization[J] . The Scientific World Journal, 2014, 2014(5):articleID 625342.
[9] 闫健卓, 李鹏英, 方丽英, 等. 基于χ2 统计的改进文本特征选择方法[J] . 计算机工程与设计, 2016, 37(5):1391-1394.
[10] 任永功, 杨雪, 杨荣杰, 等. 基于信息增益特征关联树的文本特征选择算法[J] . 计算机科学, 2013, 40(10):252-256.
[11] 尤鸣宇, 陈燕, 李国正. 不均衡问题中的特征选择新算法:Im-IG[J] . 山东大学学报:工学版, 2010, 40(5):123-128.
[12] 徐燕, 李锦涛, 王斌, 等. 不均衡数据集上文本分类的特征选择研究[J] . 计算机研究与发展, 2007, 44(S1):58-62.
[13] 张玉芳, 王勇, 熊忠阳, 等. 不平衡数据集上的文本分类特征选择新方法[J] . 计算机应用研究, 2011, 28(12):4532-4534.
[14] Uysal A K, Gunal S. A novel probabilistic feature selection method for text classification[J] . Knowledge-Based Systems, 2012, 36(12):226-235.
[15] Ogura H, Amano H, Kondo M. Comparison of metrics for feature selection in imbalanced text classification[J] . Expert Systems with Applications, 2011, 38(5):4978-4989.
[16] Fausett LV. Fundamentals of neural networks:architectures, algorithms, and applications[M] . New Jersey:Prentice-Hall Inc, 1994.
[17] Pietramala A, Policicchio V L, Rullo P. Automatic filtering of valuable features for text categorization[C] // Proc of International Conference on Advanced Data Mining and Applications. Berlin:Springer, 2012:284-295.
收稿日期 2017/3/21
修回日期 2017/5/16
页码 1947-1949,1969
中图分类号 TP391.1
文献标志码 A