广东工业大学学报 ›› 2019, Vol. 36 ›› Issue (05): 7-13.doi: 10.12052/gdutxb.190048

• 综合研究 • 上一篇    下一篇

联合低秩表示与图嵌入的无监督特征选择

滕少华1, 冯镇业1, 滕璐瑶2, 房小兆1   

  1. 1. 广东工业大学 计算机学院, 广东 广州 510006;
    2. 维多利亚大学 应用信息中心, 维多利亚州 墨尔本 VIC 3011
  • 收稿日期:2019-03-25 出版日期:2019-08-21 发布日期:2019-08-06
  • 作者简介:滕少华(1962-),男,教授,博士,主要研究方向为大数据、角色分配、网络安全、协同计算、机器学习.
  • 基金资助:
    国家自然科学基金资助项目（61702110，61772141）；广东省教育厅项目（粤教高函〔2018〕179号）；广州市科技计划项目（201802010042）

Joint Low-Rank Representation and Graph Embedding for Unsupervised Feature Selection

Teng Shao-hua1, Feng Zhen-ye1, Teng Lu-yao2, Fang Xiao-zhao1   

  1. 1. School of Computers, Guangdong University of Technology, Guangzhou 510006, China;
    2. Centre for Applied Informatics, Victoria University, Melbourne VIC 3011, Australia
  • Received:2019-03-25 Online:2019-08-21 Published:2019-08-06

HTML

PDF (PC)

2707

摘要： 大数据应用带来高维数据急剧增加，数据降维已成为重要问题.特征选择降维方法已广泛应用于模式识别领域，近年来提出了许多基于流形学习的特征选择方法，然而这类方法往往容易受到各种噪声影响.对此，本文提出一种联合低秩表示和图嵌入的高效无监督特征选择方法（JLRRGE）.通过低秩表示寻找数据在低秩子空间下的表示，降低噪声的影响从而提高算法的鲁棒性，并通过自适应图嵌入方法，使选择特征保持原有的局部关系.实验结果表明，本文提出算法的分类准确率优于其他对比算法.

关键词: 无监督学习, 低秩表示, 图嵌入, 特征选择

Abstract: Dimensionality reduction becomes a significant problem due to the proliferation of high dimensional data. For dimensionality reduction, feature selection is more analytical than feature extraction. Therefore feature selection plays an important role in pattern recognition. In recent years, many feature selection methods based on manifold learning have been proposed. However, such methods are susceptible to noise data. Therefore, an efficient unsupervised feature selection method is proposed-joint low-rank representation and graph embedding for unsupervised feature selection (JLRRGE). This method not only finds the low rank structure of the data after selecting feature, which makes the algorithm more robust, but also preserves the local manifold structure of the data through the adaptive graph embedding. The experimental results show that the proposed algorithm is superior to other compared methods.

Key words: unsupervised learning, low rank representation, graph embedding, feature selection

中图分类号: 

  • TP391.4
[1] XU Y, FANG X Z, LI X L, <i>et al</i>. Data uncertainty in face recognition[J]. IEEE Transactions on Cybernetics, 2014, 44(10):1950-1961<br /> [2] YAO C, CHENG G. Approximative bayes optimality linear discriminant analysis for Chinese handwriting character recognition[J]. Neuro-computing, 2016, 207:346-353<br /> [3] FEI L K, ZHANG B, ZHANG W, <i>et al</i>. Local apparent and latent direction extraction for palm-print recognition[J]. Information Sciences, 2019, 473:59-72<br /> [4] LAN X, MA A J, YUEN P C, <i>et al</i>. Joint sparse representation and robust feature-level fusion for multicue visual tracking[J]. IEEE Transactions on Image Processing, 2015, 24(12):5826-5841<br /> [5] FANG X Z, HAN N, WU J G, <i>et al</i>. Approximate low-rank projection learning for feature extraction[J]. IEEE Transactions on Neural Networks and Learning Systems, 2018, 29(11):5228-5241<br /> [6] 滕少华, 郑明, 刘冬宁. 面向音乐推荐的全变差图非负矩阵分解方法[J]. 计算机应用研究, 2018, 35(4):1010-1013 TENG S H, ZHENG M, LIU D L. Facing music recommended total variation non-negative matrix decomposition method[J]. Application Research of Computers, 2018, 35(4):1010-1013<br /> [7] 滕少华, 宋欢, 霍颖翔, 等. 一种增量式学习的语音字典构造方法[J]. 广东工业大学学报, 2018, 35(3):29-36 TENG S H, SONG H, HUO Y X, <i>et al</i>. An incremental learning approach in voice compression via sparse dictionary learning[J]. Journal of Guangdong University of Technology, 2018, 35(3):29-36<br /> [8] JA IN, DU IN, MAO J. Statistical pattern recognition:a review[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2002, 27(11):1502-1502<br /> [9] FANG X Z, XU Y, LI X L, <i>et al</i>. Locality and similarity preserving embedding for feature selection[J]. Neurocomputing, 2014, 128:304-315<br /> [10] 滕少华, 卢东略, 霍颖翔, 等. 基于正交投影的降维分类方法研究[J]. 广东工业大学学报, 2017, 34(3):1-7 TENG S H, LU D L, HUO Y X, <i>et al</i>. Classification method based on dimension reduction[J]. Journal of Guangdong University of Technology, 2017, 34(3):1-7<br /> [11] IMOTO S, MIVANO S. A top-r feature selection algorithm for microarray gene data[J]. IEEE/ACM Transactions on Computational Bi-ology and Bioinformatics, 2012, 9(3):754-764<br /> [12] SHANG C, LI M, FENG S, <i>et al</i>. Feature selection via maximizing global information gain for text classification[J]. Knowledge-Based Systems, 2013, 54:298-309<br /> [13] NIE F P, HUANG H, CAI X, et al. Efficient and robust feature selection via joint ?2, 1-norms minimization[C]//Advances in Neural Information Processing Systems 23:24th Annual Conference on Neural Information Processing Systems 2010. Proceedings of a meeting held 6-9 December 2010, Vancouver, British Columbia, Canada. Curran Associates Inc. 2010.<br /> [14] FANG X Z, XU Y, LI X, <i>et al</i>. Robust semi-supervised subspace clustering via non-negative low-rank representation[J]. IEEE Transactions on Cybernetics, 2016, 46(8):1828-1838<br /> [15] ZHU X F, LI X L, <i>et al</i>. Robust joint graph sparse coding for unsupervised spectral feature selection[J]. IEEE Transactions on Neural Networks and Learning Systems, 2017, 6(28):1263-1275<br /> [16] CHENG Q, ZHOU H, CHENG J. The fisher-markov selector:fast selecting maximally separable feature subset for multiclass classification with applications to high-dimensional data[J]. IEEE Transactions on Software Engineering, 2011, 33(6):1217-1233<br /> [17] CAI D, ZHANG C, HE X. Unsupervised feature selection for multi-cluster data[C]//ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. USA, Washington, DC:ACM, 2010.<br /> [18] NIE F P, XU D, TSANG W H, <i>et al</i>. Flexible manifold embedding:a framework for semi-supervised and unsupervised dimension reduction[J]. IEEE Transactions on Image Processing, 2010, 19(7):1921-1932<br /> [19] HOU C P, NIE F P, LI X L, <i>et al</i>. Joint embedding learning and sparse regression:a framework for unsupervised feature selection[J]. IEEE Transactions on Cybernetics, 2013, 44(6):793-804<br /> [20] LI Z, YANG Y, LIU J, et al. Unsupervised feature selection using nonnegative spectral analysis[C]//Proceedings of 26th AAAI Conference on Artificial Intelligence. Toronto:AAAI Press, 2012:1026-1032.<br /> [21] QIAN M, ZHAI C. Robust unsupervised feature selection[C]//International Joint Conference on Artificial Intelligence. China, Beijing:ACM 2013.<br /> [22] SHI L, DU L, SHEN Y D. Robust spectral learning for unsupervised feature selection[C]//IEEE International Conference on Data Mining. China, Shenzhen:IEEE, 2015.<br /> [23] NIE F P, ZHU W, Li X L. Unsupervised feature selection with structured graph optimization[C]//Thirtieth Aaai Conference on Artificial Intelligence. USA, Phoenix:AAAI Press, 2016.<br /> [24] WEN J, HAN N, FANG X Z, <i>et al</i>. Low-rank preserving projection Via graph regularized reconstruction[J]. IEEE Transactions on Cybernetics, 2018:1-13<br /> [25] CANDES E J, RECHT B. Exact matrix completion via convex optimization[J]. Foundations of Computational Mathematics, 2009, 9(6):717
[1] 张巍, 张圳彬. 联合图嵌入与特征加权的无监督特征选择[J]. 广东工业大学学报, 2021, 38(05): 16-23.
[2] 谭有新, 滕少华. 短文本特征的组合加权方法[J]. 广东工业大学学报, 2020, 37(05): 51-61.
[3] 岑仕杰, 何元烈, 陈小聪. 结合注意力与无监督深度学习的单目深度估计[J]. 广东工业大学学报, 2020, 37(04): 35-41.
[4] 陈平华, 黄辉, 麦淼, 周宏虹. 结合ReliefF和互信息的多标签特征选择算法[J]. 广东工业大学学报, 2018, 35(05): 20-25,50.
[5] 贺科达, 朱铮涛, 程昱. 基于改进TF-IDF算法的文本分类方法研究[J]. 广东工业大学学报, 2016, 33(05): 49-53.
[6] 蒋盛益, 王连喜. 聚类分析研究的挑战性问题[J]. 广东工业大学学报, 2014, 31(3): 32-38.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
版权所有 © 2017 《广东工业大学学报》编辑部
地址：广州市东风东路729号广东工业大学 邮编：510090
电话：020-37626653,020-37626139,020-37626396
邮箱：xbzrb@gdut.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发