Interpretable Discriminative Dimensionality Reduction and Feature Selection on the Manifold

被引:8
|
作者
Hosseini, Babak [1 ]
Hammer, Barbara [1 ]
机构
[1] Bielefeld Univ, CITEC Ctr Excellence, Bielefeld, Germany
关键词
Dimensionality reduction; Interpretability; Supervised;
D O I
10.1007/978-3-030-46150-8_19
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Dimensionality reduction (DR) on the manifold includes effective methods which project the data from an implicit relational space onto a vectorial space. Regardless of the achievements in this area, these algorithms suffer from the lack of interpretation of the projection dimensions. Therefore, it is often difficult to explain the physical meaning behind the embedding dimensions. In this research, we propose the interpretable kernel DR algorithm (I-KDR) as a new algorithm which maps the data from the feature space to a lower dimensional space where the classes are more condensed with less overlapping. Besides, the algorithm creates the dimensions upon local contributions of the data samples, which makes it easier to interpret them by class labels. Additionally, we efficiently fuse the DR with feature selection task to select the most relevant features of the original space to the discriminative objective. Based on the empirical evidence, I-KDR provides better interpretations for embedding dimensions as well as higher discriminative performance in the embedded space compared to the state-of-the-art and popular DR algorithms.
引用
收藏
页码:310 / 326
页数:17
相关论文
共 50 条
  • [1] Feature selection for dimensionality reduction
    Mladenic, Dunja
    SUBSPACE, LATENT STRUCTURE AND FEATURE SELECTION, 2006, 3940 : 84 - 102
  • [2] Discriminative structure discovery via dimensionality reduction for facial image manifold
    Xiaoming Chen
    Ke Fan
    Wanquan Liu
    Xin Zhang
    Mingliang Xue
    Neural Computing and Applications, 2015, 26 : 373 - 381
  • [3] Discriminative structure discovery via dimensionality reduction for facial image manifold
    Chen, Xiaoming
    Fan, Ke
    Liu, Wanquan
    Zhang, Xin
    Xue, Mingliang
    NEURAL COMPUTING & APPLICATIONS, 2015, 26 (02): : 373 - 381
  • [4] Feature Selection and Dimensionality Reduction on Gene Expressions
    Kaya, Mahmut
    Bilge, Hasan Sakir
    Yildiz, Oktay
    2013 21ST SIGNAL PROCESSING AND COMMUNICATIONS APPLICATIONS CONFERENCE (SIU), 2013,
  • [5] Efricient dimensionality reduction approaches for feature selection
    Deisy, C.
    Subbulakshmi, B.
    Baskar, S.
    Ramaraj, N.
    ICCIMA 2007: INTERNATIONAL CONFERENCE ON COMPUTATIONAL INTELLIGENCE AND MULTIMEDIA APPLICATIONS, VOL II, PROCEEDINGS, 2007, : 121 - +
  • [6] Discriminative Semi-Supervised Feature Selection via Manifold Regularization
    Xu, Zenglin
    Jin, Rong
    Lyu, Michael R.
    King, Irwin
    21ST INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE (IJCAI-09), PROCEEDINGS, 2009, : 1303 - 1308
  • [7] Manifold regularized discriminative feature selection for multi-label learning
    Zhang, Jia
    Luo, Zhiming
    Li, Candong
    Zhou, Changen
    Li, Shaozi
    PATTERN RECOGNITION, 2019, 95 : 136 - 150
  • [8] Discriminative Feature Selection by Optimal Manifold Search for Neoplastic Image Recognition
    Itoh, Hayato
    Mori, Yuichi
    Misawa, Masashi
    Oda, Masahiro
    Kudo, Shin-Ei
    Mori, Kensaku
    COMPUTER VISION - ECCV 2018 WORKSHOPS, PT IV, 2019, 11132 : 534 - 549
  • [9] Discriminative Semi-Supervised Feature Selection Via Manifold Regularization
    Xu, Zenglin
    King, Irwin
    Lyu, Michael Rung-Tsong
    Jin, Rong
    IEEE TRANSACTIONS ON NEURAL NETWORKS, 2010, 21 (07): : 1033 - 1047
  • [10] Automatic Dimensionality Estimation for Manifold Learning through Optimal Feature Selection
    Dornaika, Fadi
    Assoum, Ammar
    Raducanu, Bogdan
    STRUCTURAL, SYNTACTIC, AND STATISTICAL PATTERN RECOGNITION, 2012, 7626 : 575 - 583