面向不平衡文本情感分类的三支决策特征选择方法

doi:10.11772/j.issn.1001-9081.2019050822

计算机应用 ›› 2019, Vol. 39 ›› Issue (11): 3127-3133.DOI: 10.11772/j.issn.1001-9081.2019050822

• 2019年中国粒计算与知识发现学术会议(CGCKD2019)论文 • 上一篇下一篇

面向不平衡文本情感分类的三支决策特征选择方法

万志超¹, 胡峰^1,2, 邓维斌²

1. 重庆邮电大学计算机科学与技术学院, 重庆 400065;
2. 计算智能重庆市重点实验室(重庆邮电大学), 重庆 400065

收稿日期:2019-05-06 修回日期:2019-05-23 出版日期:2019-11-10 发布日期:2019-09-11
通讯作者: 万志超
作者简介:万志超(1995-),男,湖北汉川人,硕士研究生,主要研究方向:自然语言处理、机器学习;胡峰(1978-),男,湖北天门人,教授,博士,主要研究方向:粗糙集、数据挖掘;邓维斌(1978-),男,重庆人,教授,博士,主要研究方向:不确定性决策方法。
基金资助:
国家重点研发计划项目（2018YFC0832100，2018YFC0832102）；国家自然科学基金资助项目（61533020，61751312，61309014）；重庆市基础科学与前沿技术研究专项（cstc2017jcyjAX0408）。

Feature selection method for imbalanced text sentiment classification based on three-way decisions

WAN Zhichao¹, HU Feng^1,2, DENG Weibin²

1. College of Computer Science and Technology, Chongqing University of Posts and Telecommunications, Chongqing 400065, China;
2. Chongqing Key Laboratory of Computational Intelligence(Chongqing University of Posts and Telecommunications), Chongqing 400065, China

Received:2019-05-06 Revised:2019-05-23 Online:2019-11-10 Published:2019-09-11
Supported by:
This work is partially supported by the National Key Research and Development Program of China (2018YFC0832100, 2018YFC0832102), the National Natural Science Foundation of China (61751312, 61533020, 61309014), the Chongqing Research Program of Basic Research and Frontier Technology (cstc2017jcyjAX0408).

摘要/Abstract

摘要： 传统的特征选择方法在面对不平衡文本情感倾向性分类时会有很大的局限性，这种局限性主要体现在特征维数过高、特征过于稀疏和特征分布不平衡，这会使得分类的准确度大幅度下降。根据不平衡文本情感特征分布的特点，结合三支决策的思想，提出了一种面向不平衡文本情感分类的三支决策特征选择方法（TWD-FS）。该方法将两种有监督特征选择方法相结合，将选择出的特征词进一步筛选，使得最终选择出的特征词同时满足类间离散度最大和类内离散度最小的特点，有效地减少了特征词的数量，降低了特征维度；此外，通过组合正负类情感特征，缓解了情感特征的不平衡性，有效提高了不平衡样本中少数类情感的分类效果。在COAE2013中文微博非平衡数据集等多个数据集上的实验结果表明，所提的特征选择算法TWD-FS可以有效提高不平衡文本情感分类的准确度。

关键词: 不平衡文本, 特征选择, 情感分类, 有监督, 三支决策

Abstract: Traditional feature selection methods have great limitations in the imbalanced text sentiment tendency classification, which are mainly reflected in the high feature dimension, the sparse characteristics, and the imbalanced feature distribution, making the reduction of classification accuracy. According to the distribution of emotional features of imbalanced texts, a Three-Way Decisions-Feature Selection algorithm (TWD-FS) was proposed for imbalanced text sentiment classification based on three-way decisions. In order to reduce the number of feature words and reduce the feature dimension, two supervised feature selection methods were combined, and the feature words selected were further filtered in order to make them satisfy the characteristics of the maximum between-class scatter degree and the minimum within-class scatter degree. In addition, the imbalance of sentiment features was decreased and the classification accuracy of minority sentiment was effectively improved by combining positive and negative sentiment features. The experimental results on COAE2013 Chinese microblog imbalanced datasets and other datasets show that the proposed feature selection algorithm TWD-FS can effectively improve the accuracy of imbalanced text sentiment classification.

Key words: imbalanced text, feature selection, sentiment classification, supervised, three-way decisions

中图分类号:

TP391.1

万志超, 胡峰, 邓维斌. 面向不平衡文本情感分类的三支决策特征选择方法[J]. 计算机应用, 2019, 39(11): 3127-3133.

WAN Zhichao, HU Feng, DENG Weibin. Feature selection method for imbalanced text sentiment classification based on three-way decisions[J]. Journal of Computer Applications, 2019, 39(11): 3127-3133.

参考文献

[1] 赵立东. 面向文本情感分类的非平衡数据采样方法研究[D]. 太原:山西大学, 2013. (ZHAO L D. Research on imbalanced data sampling methods for text sentiment classification[D]. Taiyuan:Shanxi University, 2013.)
[2] 田锋, 兰田, CHAO Kuo-Ming, 等. 领域实例迁移的交互文本非平衡情感分类方法[J]. 西安交通大学学报, 2015, 49(4):67-72. (TIAN F, LAN T, CHAO K-M, et al. An unbalanced emotion classification method foe interactive texts based on multiple-domain instance transfer[J]. Journal of Xi'an Jiaotong University, 2015, 49(4):67-72.)
[3] 王中卿. 基于不平衡数据的情感分类方法研究[D]. 苏州:苏州大学,2012. (WANG Z Q. Research on sentiment classification based-upon imbalanced data[D]. Soochow:Soochow University, 2012.)
[4] 王杰, 李德玉, 王素格. 面向非平衡文本情感分类的TSF特征选择方法[J]. 计算机科学, 2016, 43(10):206-210. (WANG J, LI D Y, WANG S G. TSF feature selection method for imbalanced text sentiment classification[J]. Computer Science, 2016, 43(10):206-210.)
[5] WASIKOWSKI M, CHEN X. Combating the small sample class imbalance problem using feature selection[J]. IEEE Transactions on Knowledge & Data Engineering, 2010, 22(10):1388-1400.
[6] YIN L, GE Y, XIAO K. et al. Feature selection for high-dimensional imbalanced data[J]. Neurocomputing, 2013, 105(4):3-11.
[7] YU H, WANG X, WANG G. A semi-supervised three-way clustering framework for multi-view data[C]//IJCRS 2017:International Joint Conference on Rough Sets. Berlin:Springer, 2017:313-325.
[8] 张梅山,邓知龙,车万翔,等.统计与词典相结合的领域自适应中文分词[J].中文信息学报, 2012, 26(2):8-12. (ZHANG M S, DENG Z L, CHE W X,et al. Combining statistical model and dictionary for domain adaption of Chinese word segmentation[J]. Journal of Chinese Information Processing, 2012, 26(2):8-12.)
[9] 史庆伟, 从世源, 唐晓亮. LSI_LDA:一种混合特征降维方法[J]. 计算机应用研究, 2017, 34(8):2269-2273.(SHI Q W, CONG S Y, TANG X L. LSI_LDA:mixture method for feature dimensionality reduction[J]. Application Research of Computers, 2017, 34(8):2269-2273.)
[10] AL SHAMSI F, AUNG Z. Automatic patent classification by a three-phase model with document frequency matrix and boosted tree[C]//Proceedings of the 2016 5th International Conference on Electronic Devices, Systems and Applications. Piscataway:IEEE, 2016:1-4.
[11] IBRAHIM O A S, LANDA-SILVA D. Term frequency with average term occurrences for textual information retrieval[J]. Soft Computing, 2016, 20(8):3045-3061.
[12] CHEN K, ZHANG Z, LONG J, et al. Turning from TF-IDF to TF-IGM for term weighting in text classification[J]. Expert Systems with Applications:an International Journal, 2016, 66(C):245-260.
[13] 毛临川, 吴根秀, 吴恒, 等. 基于信息增益的最优组合因子Fisher判别法[J]. 计算机工程与应用, 2016, 52(19):94-96. (MAO L C, WU G X, WU H, et al. Optimal combination of factor Fisher discrimination method based on information gain[J]. Computer Engineering and Applications, 2016, 52(19):94-96.)
[14] 李平, 戴月明, 王艳. 基于混合卡方统计量与逻辑回归的文本情感分析[J]. 计算机工程, 2017, 43(12):192-196. (LI P, DAI Y M, WANG Y. Text sentiment analysis based on hybrid chi-square statistic and logistics regression[J]. Computer Engineering, 2017, 43(12):192-196.)
[15] 段宏湘, 张秋余, 张墨逸. 基于归一化互信息的FCBF特征选择算法[J]. 华中科技大学学报(自然科学版), 2017, 45(1):52-56.(DUAN H X, ZHANG Q Y, ZHANG M Y. FCBF algorithm based on normalized mutual information for feature selection[J]. Journal of Huazhong University of Science and Technology (Natural Science Edition), 2017, 45(1):52-56.)
[16] 张辉宜, 谢业名, 袁志祥, 等. 一种基于概率的卡方特征选择方法[J]. 计算机工程, 2016, 42(8):194-198.(ZHANG H Y, XIE Y M, YUAN Z X, et al. A method of CHI-square feature selection based on probability[J]. Computer Engineering, 2016, 42(8):194-198.)
[17] 王晨曦, 林耀进, 刘景华, 等. 基于最近邻互信息的特征选择算法[J]. 计算机工程与应用, 2016, 52(18):74-78. (WANG C X, LIN Y J, LIU J H, et al. Feature selection algorithm based on nearest-neighbor mutual information[J]. Computer Engineering and Applications, 2016, 52(18):74-78.)
[18] 吴金源, 冀俊忠, 赵学武, 等. 基于特征选择技术的情感词权重计算[J]. 北京工业大学学报, 2016, 42(1):142-151. (WU J Y, JI J Z, ZHAO X W, et al. Weight calculation of emotional word based on feature selection technique[J]. Journal of Beijing University of Technology, 2016, 42(1):142-151.)
[19] YAO Y. The superiority of three-way decisions in probabilistic rough set models[J]. Information Sciences, 2011, 181(6):1080-1096.
[20] 姚海英. 中文文本分类中卡方统计特征选择方法和TF-IDF权重计算方法的研究[D].长春:吉林大学, 2016. (YAO H Y. Research on chi-square statistic feature selection method and TF-IDF feature weighting method for chinese text classification[D]. Changchun:Jilin University, 2016.)
[21] 李燕, 卫志华, 徐凯. 基于Lasso算法的中文情感混合特征选择方法研究[J]. 计算机科学, 2018, 45(1):39-46. (LI Y, WEI Z H, XU K. Hybrid feature selection method of chinese emotional characteristics based on Lasso algorithm[J]. Computer Science, 2018, 45(1):39-46.)
[22] 张越兵,苗夺谦,张志飞.基于三支决策的多粒度文本情感分类模型[J].计算机科学, 2017, 44(12):188-193. (ZHANG Y B, MIAO D Q, ZHANG Z F. Multi-granularity text sentiment classification model based on three-way decisions[J]. Computer Science, 2017, 44(12):188-193.)

面向不平衡文本情感分类的三支决策特征选择方法

Feature selection method for imbalanced text sentiment classification based on three-way decisions

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	张师鹏, 李永忠, 杜祥通. 基于半监督学习和三支决策的入侵检测模型[J]. 计算机应用, 2021, 41(9): 2602-2608.
[2]	湛航, 何朗, 黄樟灿, 李华峰, 张蔷, 谈庆. 改进的基于层次距离的基因表达式编程特征选择分类算法[J]. 计算机应用, 2021, 41(9): 2658-2667.
[3]	祝承, 赵晓琦, 赵丽萍, 焦玉宏, 朱亚飞, 陈建英, 周伟, 谭颖. 基于谱聚类半监督特征选择的功能磁共振成像数据分类[J]. 计算机应用, 2021, 41(8): 2288-2293.
[4]	李蒙蒙, 秦伟, 刘艺, 刁兴春. 结合头脑风暴优化的混合蚁群优化算法[J]. 计算机应用, 2021, 41(8): 2412-2417.
[5]	贾鹤鸣, 姜子超, 李瑶, 孙康健. 基于改进斑点鬣狗优化算法的同步优化特征选择[J]. 计算机应用, 2021, 41(5): 1290-1298.
[6]	林筠超, 万源. 基于图结构优化的自适应多度量非监督特征选择方法[J]. 计算机应用, 2021, 41(5): 1282-1289.
[7]	杨翎, 姜春茂. 基于三支决策的虚拟机节能迁移策略[J]. 计算机应用, 2021, 41(4): 990-998.
[8]	袁景凌, 丁远远, 潘东行, 李琳. 基于时序和上下文特征的中文隐式情感分类模型[J]. 计算机应用, 2021, 41(10): 2820-2828.
[9]	张志浩, 林耀进, 卢舜, 郭晨, 王晨曦. 缺失标记下基于类属属性的多标记特征选择[J]. 计算机应用, 2021, 41(10): 2849-2857.
[10]	罗俊, 陈黎飞. 基于BERT的不完全数据情感分类[J]. 计算机应用, 2021, 41(1): 139-144.
[11]	顾桐, 许国良, 李万林, 李家浩, 王志愿, 雒江涛. 基于集成LightGBM和贝叶斯优化策略的房价智能评估模型[J]. 计算机应用, 2020, 40(9): 2762-2767.
[12]	尹春勇, 章荪. 面向短文本情感分类的端到端对抗变分贝叶斯方法[J]. 计算机应用, 2020, 40(9): 2536-2542.
[13]	黄学雨, 徐浩特, 陶剑文. 具有特征选择的多源自适应分类框架[J]. 计算机应用, 2020, 40(9): 2499-2506.
[14]	肖跃雷, 张云娇. 基于特征选择和超参数优化的恐怖袭击组织预测方法[J]. 计算机应用, 2020, 40(8): 2262-2267.
[15]	刘丹, 姚立霜, 王云锋, 裴作飞. 面向类不平衡流量数据的分类模型[J]. 计算机应用, 2020, 40(8): 2327-2333.