Towards Ultrahigh Dimensional Feature Selection for Big Data

被引:0
|
作者
Tan, Mingkui [1 ]
Tsang, Ivor W. [2 ]
Wang, Li [3 ]
机构
[1] Nanyang Technol Univ, Sch Comp Engn, Singapore 639798, Singapore
[2] Univ Technol Sydney, Ctr Quantum Computat & Intelligent Syst, Broadway, NSW 2007, Australia
[3] Univ Calif San Diego, Dept Math, La Jolla, CA 92093 USA
基金
澳大利亚研究理事会;
关键词
big data; ultrahigh dimensionality; feature selection; nonlinear feature selection; multiple kernel learning; feature generation; MULTIPLE; CLASSIFICATION; OPTIMIZATION; CONVERGENCE; ONLINE; CANCER; LASSO;
D O I
暂无
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
In this paper, we present a new adaptive feature scaling scheme for ultrahigh-dimensional feature selection on Big Data, and then reformulate it as a convex semi-infinite programming (SIP) problem. To address the SIP, we propose an efficient feature generating paradigm. Different from traditional gradient-based approaches that conduct optimization on all input features, the proposed paradigm iteratively activates a group of features, and solves a sequence of multiple kernel learning (MKL) subproblems. To further speed up the training, we propose to solve the MKL subproblems in their primal forms through a modified accelerated proximal gradient approach. Due to such optimization scheme, some efficient cache techniques are also developed. The feature generating paradigm is guaranteed to converge globally under mild conditions, and can achieve lower feature selection bias. Moreover, the proposed method can tackle two challenging tasks in feature selection: 1) group-based feature selection with complex structures, and 2) nonlinear feature selection with explicit feature mappings. Comprehensive experiments on a wide range of synthetic and real-world data sets of tens of million data points with O(10(14)) features demonstrate the competitive performance of the proposed method over state-of-the-art feature selection methods in terms of generalization performance and training efficiency.
引用
收藏
页码:1371 / 1429
页数:59
相关论文
共 50 条
  • [21] An online approach for feature selection for classification in big data
    Nazar, Nasrin Banu
    Senthilkumar, Radha
    TURKISH JOURNAL OF ELECTRICAL ENGINEERING AND COMPUTER SCIENCES, 2017, 25 (01) : 163 - 171
  • [22] Streaming feature selection algorithms for big data: A survey
    AlNuaimi, Noura
    Masud, Mohammad Mehedy
    Serhani, Mohamed Adel
    Zaki, Nazar
    APPLIED COMPUTING AND INFORMATICS, 2022, 18 (1/2) : 113 - 135
  • [23] Scalable and Accurate Online Feature Selection for Big Data
    Yu, Kui
    Wu, Xindong
    Ding, Wei
    Pei, Jian
    ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM DATA, 2016, 11 (02)
  • [24] Ensemble with Divisive Bagging for Feature Selection in Big Data
    Park, Yousung
    Kwon, Tae Yeon
    COMPUTATIONAL ECONOMICS, 2024,
  • [25] Distributed Evolutionary Feature Selection for Big Data Processing
    Bouaguel, Waad
    Ben NCir, Chiheb Eddine
    VIETNAM JOURNAL OF COMPUTER SCIENCE, 2022, 09 (03) : 313 - 332
  • [26] Improved Feature Selection Model for Big Data Analytics
    El-Hasnony, Ibrahim M.
    Barakat, Sherif I.
    Elhoseny, Mohamed
    Mostafa, Reham R.
    IEEE ACCESS, 2020, 8 : 66989 - 67004
  • [27] Reducing Data Complexity in Feature Extraction and Feature Selection for Big Data Security Analytics
    Sisiaridis, Dimitrios
    Markowitch, Olivier
    2018 1ST INTERNATIONAL CONFERENCE ON DATA INTELLIGENCE AND SECURITY (ICDIS 2018), 2018, : 43 - 48
  • [28] Feature Selection for Big Visual Data: Overview and Challenges
    Bolon-Canedo, Veronica
    Remeseiro, Beatriz
    Cancela, Brais
    IMAGE ANALYSIS AND RECOGNITION (ICIAR 2018), 2018, 10882 : 136 - 143
  • [29] Feature Selection for Varying Coefficient Models With Ultrahigh-Dimensional Covariates
    Liu, Jingyuan
    Li, Runze
    Wu, Rongling
    JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 2014, 109 (505) : 266 - 274
  • [30] A selective overview of feature screening for ultrahigh-dimensional data
    LIU JingYuan
    ZHONG Wei
    LI RunZe
    Science China(Mathematics), 2015, 58 (10) : 2033 - 2054