Weighted Neighborhood Preserving Ensemble Embedding

被引:7
|
作者
Mehta, Sumet [1 ,2 ]
Zhan, Bi-Sheng [1 ]
Shen, Xiang-Jun [1 ]
机构
[1] Jiangsu Univ, Sch Comp Sci & Commun Engn, Zhenjiang 212013, Jiangsu, Peoples R China
[2] JCDM Coll Engn, Dept Elect & Commun Engn, Sirsa 125055, Haryana, India
来源
ELECTRONICS | 2019年 / 8卷 / 02期
基金
中国国家自然科学基金;
关键词
dimensional reduction; graph construction; machine learning; neighborhood preserving embedding; face recognition; DIMENSIONALITY REDUCTION; FACE RECOGNITION; ILLUMINATION; PROJECTIONS; EIGENFACES; ALGORITHM;
D O I
10.3390/electronics8020219
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Neighborhood preserving embedding (NPE) is a classical and very promising supervised dimensional reduction (DR) technique based on a linear graph, which preserves the local neighborhood relations of the data points. However, NPE uses the K nearest neighbor (KNN) criteria for constructing an adjacent graph which makes it more sensitive to neighborhood size. In this article, we propose a novel DR method called weighted neighborhood preserving ensemble embedding (WNPEE). Unlike NPE, the proposed WNPEE constructs an ensemble of adjacent graphs with the number of nearest neighbors varying. With this graph ensemble building, WNPEE can obtain the low-dimensional projections with optimal embedded graph pursuing in a joint optimization manner. WNPEE can be applied in many machine learning fields, such as object recognition, data classification, signal processing, text categorization, and various deep learning tasks. Extensive experiments on Olivetti Research Laboratory (ORL), Georgia Tech, Carnegie Mellon University-Pose and Illumination Images (CMU PIE) and Yale, four face databases demonstrate that WNPEE achieves a competitive and better recognition rate than NPE and other comparative DR methods. Additionally, the proposed WNPEE achieves much lower sensitivity to the neighborhood size parameter as compared to the traditional NPE method while preserving more of the local manifold structure of the high-dimensional data.
引用
收藏
页数:18
相关论文
共 50 条
  • [1] Neighborhood preserving embedding
    He, XF
    Cai, D
    Yan, SC
    Zhang, HJ
    TENTH IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION, VOLS 1 AND 2, PROCEEDINGS, 2005, : 1208 - 1213
  • [2] Input Decimated Ensemble based on Neighborhood Preserving Embedding for spectrogram classification
    Nanni, Loris
    Lumini, Alessandra
    EXPERT SYSTEMS WITH APPLICATIONS, 2009, 36 (08) : 11257 - 11261
  • [3] Stable neighborhood preserving embedding
    Wang, Jing
    CIS WORKSHOPS 2007: INTERNATIONAL CONFERENCE ON COMPUTATIONAL INTELLIGENCE AND SECURITY WORKSHOPS, 2007, : 252 - 255
  • [4] Neighborhood preserving embedding with autoencoder
    Rana, Ruisheng
    Wang, Jinping
    Fang, Bin
    Yang, Weiming
    DIGITAL SIGNAL PROCESSING, 2024, 145
  • [5] Joint Sparse Neighborhood Preserving Embedding
    Liu, Haibiao
    Lai, Zhihui
    Chen, Yudong
    2018 INTERNATIONAL SEMINAR ON COMPUTER SCIENCE AND ENGINEERING TECHNOLOGY (SCSET 2018), 2019, 1176
  • [6] Quantum algorithm for neighborhood preserving embedding
    Pan, Shi-Jie
    Wan, Lin-Chun
    Liu, Hai-Ling
    Wu, Yu-Sen
    Qin, Su-Juan
    Wen, Qiao-Yan
    Gao, Fei
    CHINESE PHYSICS B, 2022, 31 (06)
  • [7] KERNEL NEIGHBORHOOD PRESERVING EMBEDDING FOR CLASSIFICATION
    Tao Xiaoyan Ji Hongbing Men Jian(School of Electronic Engineering
    JournalofElectronics(China), 2009, 26 (03) : 374 - 379
  • [8] Constraint Sparse Neighborhood Preserving Embedding
    Xu, Shuhua
    Open Cybernetics and Systemics Journal, 2014, 8 (01): : 208 - 216
  • [9] Flexible Orthogonal Neighborhood Preserving Embedding
    Pang, Tianji
    Nie, Feiping
    Han, Junwei
    PROCEEDINGS OF THE TWENTY-SIXTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, 2017, : 2592 - 2598
  • [10] Analysis on Supervised Neighborhood Preserving Embedding
    Teoh, Andrew B. J.
    Pang, Ying Han
    IEICE ELECTRONICS EXPRESS, 2009, 6 (23): : 1631 - 1637