Feature selection based on sensitivity analysis of fuzzy ISODATA

被引：24
作者：
Liu, Quanjin [1 ,2 ]
Zhao, Zhimin [1 ]
Li, Ying-Xin [3 ]
Li, Yuanyuan [1 ]
机构：
[1] Nanjing Univ Aeronaut & Astronaut, Coll Sci, Nanjing 210016, Jiangsu, Peoples R China
[2] AnQing Normal Coll, Sch Phys & Elect Engn, Anqing 246011, Peoples R China
[3] Beijing Jingwei Text Machinery New Technol Co Ltd, Inst Machine Vis & Machine Intelligence, Beijing 100176, Peoples R China
来源：
NEUROCOMPUTING | 2012年 / 85卷
基金：
高等学校博士学科点专项科研基金; 中国国家自然科学基金;
关键词：
Feature selection; Fuzzy ISODATA; Sensitivity analysis; Microarray; Classification; Clustering; SUPPORT VECTOR MACHINES; CANCER-DIAGNOSIS; MICROARRAY DATA; GENE SELECTION; CLASSIFICATION; ALGORITHM; PERFORMANCE; PREDICTION; WRAPPERS; TUMOR;
D O I：
10.1016/j.neucom.2012.01.005
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
A feature selection method based on sensitivity analysis and the fuzzy Interactive Self-Organizing Data Algorithm (ISODATA) is proposed for selecting features from high dimensional gene expression data sets. First, feature sensitivities for discriminating classes are calculated on the basis of the fuzzy ISODATA method. Then, candidate feature subsets are generated according to feature sensitivities with the recursive feature elimination procedure. Finally, the obtained optimal feature subsets are evaluated using both supervised and unsupervised methods to validate their abilities for separating different categories. The proposed method is applied to five microarray datasets, and the experimental results indicate its effectiveness. (c) 2012 Elsevier B.V. All rights reserved.
引用
收藏
页码：29 / 37
页数：9
相关论文
共 50 条
  • [41] Robust feature selection based on regularized brownboost loss
    Wei, Pan
    Hu, Qinghua
    Ma, Peijun
    Su, Xiaohong
    [J]. KNOWLEDGE-BASED SYSTEMS, 2013, 54 : 180 - 198
  • [42] A novel feature selection scheme for high-dimensional data sets: four-Staged Feature Selection
    Pehlivanli, Ayca Cakmak
    [J]. JOURNAL OF APPLIED STATISTICS, 2016, 43 (06) : 1140 - 1154
  • [43] A robust SVM-based approach with feature selection and outliers detection for classification problems
    Baldomero-Naranjo, Marta
    Martinez-Merino, Luisa I.
    Rodriguez-Chia, Antonio M.
    [J]. EXPERT SYSTEMS WITH APPLICATIONS, 2021, 178
  • [44] Feature selection from high dimensional data based on iterative qualitative mutual information
    Nagpal, Arpita
    Singh, Vijendra
    [J]. JOURNAL OF INTELLIGENT & FUZZY SYSTEMS, 2019, 36 (06) : 5845 - 5856
  • [45] A Feature Selection Method Based on Hybrid Improved Binary Quantum Particle Swarm Optimization
    wu, Qing
    Ma, Zheping
    Fan, Jin
    Xu, Gang
    Shen, Yuanfeng
    [J]. IEEE ACCESS, 2019, 7 : 80588 - 80601
  • [46] SVM-RFE based feature selection for tandem mass spectrum quality assessment
    Ding, Jiarui
    Shi, Jinhong
    Wu, Fang-Xiang
    [J]. INTERNATIONAL JOURNAL OF DATA MINING AND BIOINFORMATICS, 2011, 5 (01) : 73 - 88
  • [47] Unsupervised hyperspectral feature selection based on fuzzy c-means and grey wolf optimizer
    Xie, Fuding
    Lei, Cunkuan
    Li, Fangfei
    Huang, Dan
    Yang, Jun
    [J]. INTERNATIONAL JOURNAL OF REMOTE SENSING, 2019, 40 (09) : 3344 - 3367
  • [48] A fast SVM-based wrapper feature selection method driven by a fuzzy complementary criterion
    Moustakidis, S. P.
    Theocharis, J. B.
    [J]. PATTERN ANALYSIS AND APPLICATIONS, 2012, 15 (04) : 379 - 397
  • [49] The feature selection bias problem in relation to high-dimensional gene data
    Krawczuk, Jerzy
    Lukaszuk, Tomasz
    [J]. ARTIFICIAL INTELLIGENCE IN MEDICINE, 2016, 66 : 63 - 71
  • [50] Novel Multi-Class Feature Selection Methods Using Sensitivity Analysis of Posterior Probabilities
    Shen, Kai-Quan
    Ong, Chong-Jin
    Li, Xiao-Ping
    Wilder-Smith, Einar P. V.
    [J]. 2008 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN AND CYBERNETICS (SMC), VOLS 1-6, 2008, : 1115 - +
12345