On the role of sparsity in feature selection and an innovative method LRMI
Feature selection is used in many applications in machine learning and bioinformatics. As a popular approach, feature selection can be implemented in the filter-manner based on the sparse solution of the l1 regularization. Most study of the l1 regularization concentrates on investigating the iterati...
Saved in:
Published in | Neurocomputing (Amsterdam) Vol. 321; pp. 237 - 250 |
---|---|
Main Authors | , , |
Format | Journal Article |
Language | English |
Published |
Elsevier B.V
10.12.2018
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | Feature selection is used in many applications in machine learning and bioinformatics. As a popular approach, feature selection can be implemented in the filter-manner based on the sparse solution of the l1 regularization. Most study of the l1 regularization concentrates on investigating the iteration solution of the problem or focuses on adapting sparsity to different applications. It is necessary to explore more deeply about how the sparsity learned with the l1 regularization contributes to feature selection. In this paper, we make an effort to analyze the role of the l1 regularization in feature selection from the perspective of information theory. We discover that the l1 regularization contributes to minimizing the redundancy in feature selection. To avoid the complex computation of the l1 optimization, we propose a novel feature selection algorithm, i.e. the Laplacian regularization based on mutual information (LRMI), which realizes the minimization of the redundancy in a new way, and incorporates the l2 norm to achieve automatic grouping. Extensive experimental results demonstrate the superiority of LRMI over several traditional l1 regularization based feature selection algorithms with less time consumption. |
---|---|
AbstractList | Feature selection is used in many applications in machine learning and bioinformatics. As a popular approach, feature selection can be implemented in the filter-manner based on the sparse solution of the l1 regularization. Most study of the l1 regularization concentrates on investigating the iteration solution of the problem or focuses on adapting sparsity to different applications. It is necessary to explore more deeply about how the sparsity learned with the l1 regularization contributes to feature selection. In this paper, we make an effort to analyze the role of the l1 regularization in feature selection from the perspective of information theory. We discover that the l1 regularization contributes to minimizing the redundancy in feature selection. To avoid the complex computation of the l1 optimization, we propose a novel feature selection algorithm, i.e. the Laplacian regularization based on mutual information (LRMI), which realizes the minimization of the redundancy in a new way, and incorporates the l2 norm to achieve automatic grouping. Extensive experimental results demonstrate the superiority of LRMI over several traditional l1 regularization based feature selection algorithms with less time consumption. |
Author | Fang, Yuchun Yuan, Qiulong Zhang, Zhaoxiang |
Author_xml | – sequence: 1 givenname: Yuchun orcidid: 0000-0002-7085-8876 surname: Fang fullname: Fang, Yuchun email: ycfang@shu.edu.cn organization: School of Computer Engineering and Science, Shanghai University, China – sequence: 2 givenname: Qiulong surname: Yuan fullname: Yuan, Qiulong organization: School of Computer Engineering and Science, Shanghai University, China – sequence: 3 givenname: Zhaoxiang surname: Zhang fullname: Zhang, Zhaoxiang organization: National Laboratory of Pattern Recognition, CASIA, China |
BookMark | eNp9kEFLAzEQhYNUsK3-Aw_5A7tmsslu9iJIUVupFETPIc3O0pQ2Kcm20H9vSj17GOYw8x7fexMy8sEjIY_ASmBQP21Lj0cb9iVnoErWlgyaGzIG1fBCcVWPyJi1XBa8An5HJiltWf4A3o7Jx8rTYYM0hh3S0NN0MDG54Uydpz2a4RiRJtyhHVzw1PguT775cDKDOyHd47AJHV1-fS7uyW1vdgkf_vaU_Ly9fs_mxXL1vpi9LAvLpRyKzGOtqdZWguiUQSF4ZTGDcyU5imYtJIDtoBamakTNRG0Vh1aoNZieSV5Nibj62hhSitjrQ3R7E88amL70obf62oe-9KFZq3PaLHu-yjCznRxGnaxDb7FzMcfTXXD_G_wCeqRr4w |
Cites_doi | 10.1109/TPAMI.2005.55 10.1093/bioinformatics/btm344 10.1111/j.1467-9868.2011.00771.x 10.1186/1475-925X-6-23 10.1007/BF01237942 10.1109/TNNLS.2014.2341627 10.1111/j.1467-9868.2005.00527.x 10.1016/j.sigpro.2012.09.011 10.1126/science.1205438 10.1214/aoms/1177729694 10.1002/cpa.20132 10.1109/TIT.2006.871582 10.1038/35000501 10.1214/009053604000000067 10.1109/TSMCB.2011.2179533 10.1109/TPAMI.2005.159 10.1007/s00521-013-1368-0 10.1007/s11263-008-0180-2 10.1371/journal.pone.0028210 10.1016/j.patcog.2015.11.007 10.1198/016214506000000735 10.1007/BF01158386 10.1214/009053607000000875 10.1109/72.298224 10.1111/j.1467-9868.2005.00490.x 10.1109/TIP.2008.2007610 10.1109/TPAMI.2006.244 10.1109/JPROC.2010.2044470 |
ContentType | Journal Article |
Copyright | 2018 |
Copyright_xml | – notice: 2018 |
DBID | AAYXX CITATION |
DOI | 10.1016/j.neucom.2018.09.017 |
DatabaseName | CrossRef |
DatabaseTitle | CrossRef |
DatabaseTitleList | |
DeliveryMethod | fulltext_linktorsrc |
Discipline | Computer Science |
EISSN | 1872-8286 |
EndPage | 250 |
ExternalDocumentID | 10_1016_j_neucom_2018_09_017 S0925231218310786 |
GroupedDBID | --- --K --M .DC .~1 0R~ 123 1B1 1~. 1~5 4.4 457 4G. 53G 5VS 7-5 71M 8P~ 9JM 9JN AABNK AACTN AADPK AAEDT AAEDW AAIAV AAIKJ AAKOC AALRI AAOAW AAQFI AAXLA AAXUO AAYFN ABBOA ABCQJ ABFNM ABJNI ABMAC ABYKQ ACDAQ ACGFS ACRLP ACZNC ADBBV ADEZE AEBSH AEKER AENEX AFKWA AFTJW AFXIZ AGHFR AGUBO AGWIK AGYEJ AHHHB AHZHX AIALX AIEXJ AIKHN AITUG AJOXV ALMA_UNASSIGNED_HOLDINGS AMFUW AMRAJ AOUOD AXJTR BKOJK BLXMC CS3 DU5 EBS EFJIC EFLBG EJD EO8 EO9 EP2 EP3 F5P FDB FIRID FNPLU FYGXN G-Q GBLVA GBOLZ IHE J1W KOM LG9 M41 MO0 MOBAO N9A O-L O9- OAUVE OZT P-8 P-9 P2P PC. Q38 RIG ROL RPZ SDF SDG SDP SES SPC SPCBC SSN SSV SSZ T5K ZMT ~G- 29N AAQXK AAXKI AAYXX ABXDB ACNNM ADJOM ADMUD AFJKZ AKRWK ASPBG AVWKF AZFZN CITATION FEDTE FGOYB HLZ HVGLF HZ~ R2- SBC SEW WUQ XPP |
ID | FETCH-LOGICAL-c255t-286cca3bc514d8ae4423ce0182852e47b4511cd164a3746046c821948b1af0523 |
IEDL.DBID | AIKHN |
ISSN | 0925-2312 |
IngestDate | Thu Sep 26 16:07:25 EDT 2024 Fri Feb 23 02:30:26 EST 2024 |
IsPeerReviewed | true |
IsScholarly | true |
Keywords | Feature selection l1 regularization Mutual information Information theory |
Language | English |
LinkModel | DirectLink |
MergedId | FETCHMERGED-LOGICAL-c255t-286cca3bc514d8ae4423ce0182852e47b4511cd164a3746046c821948b1af0523 |
ORCID | 0000-0002-7085-8876 |
PageCount | 14 |
ParticipantIDs | crossref_primary_10_1016_j_neucom_2018_09_017 elsevier_sciencedirect_doi_10_1016_j_neucom_2018_09_017 |
PublicationCentury | 2000 |
PublicationDate | 2018-12-10 |
PublicationDateYYYYMMDD | 2018-12-10 |
PublicationDate_xml | – month: 12 year: 2018 text: 2018-12-10 day: 10 |
PublicationDecade | 2010 |
PublicationTitle | Neurocomputing (Amsterdam) |
PublicationYear | 2018 |
Publisher | Elsevier B.V |
Publisher_xml | – name: Elsevier B.V |
References | Q. Gu, Z. Li, J. Han, Generalized fisher score for feature selection (2012) arXiv preprint arXiv Donoho (bib0017) 2006; 52 Hou, Nie, Yi, Wu (bib0027) 2011 Li, Si, Zhou, Huang, Chen (bib0041) 2015; 26 Guyon, Elisseeff (bib0004) 2003; 3 Yuan, Fang (bib0016) 2017 Little, McSharry, Roberts, Costello, Moroz (bib0048) 2007; 6 M. Lichman, UCI machine learning repository, 2013, URL Li, Wang, Liu, Jiang (bib0045) 2015 Destrero, De Mol, Odone, Verri (bib0011) 2009; 83 Bruckstein, Donoho, Elad (bib0024) 2009; 51 Brown, Pocock, Zhao, LujÃ!‘n (bib0021) 2012; 13 Saeys, Inza, Larra, Aga (bib0003) 2007; 23 Reshef, Reshef, Finucane, Grossman, McVean, Turnbaugh, Lander, Mitzenmacher, Sabeti (bib0052) 2011; 334 Efron, Hastie, Johnstone, Tibshirani (bib0042) 2004; 32 Ahonen, Hadid, Pietikainen (bib0053) 2006; 28 Fang, Chang (bib0012) 2015 Yang, Moody (bib0019) 2000; 12 Huang, Horowitz, Ma (bib0031) 2008; 36 Vergara, Estvez (bib0036) 2014; 24 Destrero, De, Odone, Verri (bib0010) 2009; 18 Destrero, Mol, Odone, Verri (bib0028) 2007 . Zou (bib0029) 2006; 101 Fleuret (bib0039) 2003; 5 Wright, Ma, Mairal, Sapiro, Huang, Yan (bib0007) 2010; 98 Tibshirani (bib0040) 2011; 73 Alizadeh, Eisen, Davis, Ma, Lossos, Rosenwald, Boldrick, Sabet, Tran, Yu (bib0049) 2000; 403 Battiti (bib0038) 1994; 5 Nie, Huang, Cai, Ding (bib0013) 2010; Vol. 2 Silva, Marcal, da Silva (bib0046) 2013; Vol. 7950 Anne-Claire, Pierre, Jean-Philippe (bib0001) 2011; 6 Ng (bib0025) 2004 Aggarwal (bib0002) 2014 Xiyi Hang (bib0014) 2009; 2009 Tibshirani, Saunders, Rosset, Zhu, Knight (bib0030) 2005; 67 He, Yan, Hu, Niyogi, Zhang (bib0044) 2005; 27 Vinh, Zhou, Chan, Bailey (bib0050) 2016; 53 Zou, Hastie (bib0018) 2005; 67 Elad (bib0009) 2010 Cover, Thomas (bib0033) 2012 Cands (bib0022) 2006; 17 Donoho (bib0023) 2007; 59 Peng, Long, Ding (bib0020) 2005; 27 Gui, Sun, Ji, Tao, Tan (bib0026) 2016; PP Kullback (bib0034) 1959 Lewis (bib0037) 1992 Cheng, Liu, Yang, Chen (bib0008) 2013; 93 Zhang, Huang, Li, Metaxas (bib0032) 2012; 42 Guseinov (bib0043) 1987; 41 Guyon (bib0005) 1998; 1 Lin, Tang (bib0051) 2006 Kullback, Leibler (bib0035) 1951; 22 Yu, Fang (bib0015) 2014; 10 Elad (10.1016/j.neucom.2018.09.017_bib0009) 2010 Tibshirani (10.1016/j.neucom.2018.09.017_bib0030) 2005; 67 Zou (10.1016/j.neucom.2018.09.017_bib0018) 2005; 67 Cheng (10.1016/j.neucom.2018.09.017_bib0008) 2013; 93 10.1016/j.neucom.2018.09.017_bib0006 Xiyi Hang (10.1016/j.neucom.2018.09.017_bib0014) 2009; 2009 Anne-Claire (10.1016/j.neucom.2018.09.017_bib0001) 2011; 6 Reshef (10.1016/j.neucom.2018.09.017_bib0052) 2011; 334 Ahonen (10.1016/j.neucom.2018.09.017_bib0053) 2006; 28 Vergara (10.1016/j.neucom.2018.09.017_bib0036) 2014; 24 10.1016/j.neucom.2018.09.017_bib0047 Guyon (10.1016/j.neucom.2018.09.017_bib0005) 1998; 1 Brown (10.1016/j.neucom.2018.09.017_bib0021) 2012; 13 Lewis (10.1016/j.neucom.2018.09.017_bib0037) 1992 Guyon (10.1016/j.neucom.2018.09.017_bib0004) 2003; 3 Li (10.1016/j.neucom.2018.09.017_bib0041) 2015; 26 Saeys (10.1016/j.neucom.2018.09.017_bib0003) 2007; 23 Tibshirani (10.1016/j.neucom.2018.09.017_bib0040) 2011; 73 Hou (10.1016/j.neucom.2018.09.017_bib0027) 2011 Cover (10.1016/j.neucom.2018.09.017_bib0033) 2012 Alizadeh (10.1016/j.neucom.2018.09.017_bib0049) 2000; 403 Li (10.1016/j.neucom.2018.09.017_bib0045) 2015 Wright (10.1016/j.neucom.2018.09.017_bib0007) 2010; 98 Bruckstein (10.1016/j.neucom.2018.09.017_bib0024) 2009; 51 Guseinov (10.1016/j.neucom.2018.09.017_bib0043) 1987; 41 Zou (10.1016/j.neucom.2018.09.017_bib0029) 2006; 101 Huang (10.1016/j.neucom.2018.09.017_bib0031) 2008; 36 Yu (10.1016/j.neucom.2018.09.017_bib0015) 2014; 10 Kullback (10.1016/j.neucom.2018.09.017_bib0035) 1951; 22 Little (10.1016/j.neucom.2018.09.017_bib0048) 2007; 6 Gui (10.1016/j.neucom.2018.09.017_bib0026) 2016; PP Donoho (10.1016/j.neucom.2018.09.017_bib0023) 2007; 59 Yuan (10.1016/j.neucom.2018.09.017_bib0016) 2017 Yang (10.1016/j.neucom.2018.09.017_bib0019) 2000; 12 Ng (10.1016/j.neucom.2018.09.017_bib0025) 2004 Destrero (10.1016/j.neucom.2018.09.017_bib0010) 2009; 18 Nie (10.1016/j.neucom.2018.09.017_bib0013) 2010; Vol. 2 Fang (10.1016/j.neucom.2018.09.017_bib0012) 2015 Destrero (10.1016/j.neucom.2018.09.017_bib0028) 2007 Zhang (10.1016/j.neucom.2018.09.017_bib0032) 2012; 42 Peng (10.1016/j.neucom.2018.09.017_bib0020) 2005; 27 Efron (10.1016/j.neucom.2018.09.017_bib0042) 2004; 32 Kullback (10.1016/j.neucom.2018.09.017_bib0034) 1959 Silva (10.1016/j.neucom.2018.09.017_bib0046) 2013; Vol. 7950 Destrero (10.1016/j.neucom.2018.09.017_bib0011) 2009; 83 Vinh (10.1016/j.neucom.2018.09.017_bib0050) 2016; 53 Battiti (10.1016/j.neucom.2018.09.017_bib0038) 1994; 5 Lin (10.1016/j.neucom.2018.09.017_bib0051) 2006 Fleuret (10.1016/j.neucom.2018.09.017_bib0039) 2003; 5 Aggarwal (10.1016/j.neucom.2018.09.017_bib0002) 2014 He (10.1016/j.neucom.2018.09.017_bib0044) 2005; 27 Donoho (10.1016/j.neucom.2018.09.017_bib0017) 2006; 52 Cands (10.1016/j.neucom.2018.09.017_bib0022) 2006; 17 |
References_xml | – volume: 26 start-page: 1388 year: 2015 end-page: 1402 ident: bib0041 article-title: Frel: a stable feature selection algorithm publication-title: IEEE Trans. Neural Netw. Learn. Syst. contributor: fullname: Chen – start-page: 68 year: 2006 end-page: 82 ident: bib0051 article-title: Conditional infomax learning: an integrated framework for feature extraction and fusion publication-title: Proceedings of the European Conference on Computer Vision contributor: fullname: Tang – volume: 52 start-page: 1289 year: 2006 end-page: 1306 ident: bib0017 article-title: Compressed sensing publication-title: IEEE Trans. Inf. Theory contributor: fullname: Donoho – volume: 18 start-page: 188 year: 2009 ident: bib0010 article-title: A sparsity-enforcing method for learning face features publication-title: IEEE Trans. Image Process. contributor: fullname: Verri – start-page: 592 year: 2017 end-page: 603 ident: bib0016 article-title: A Novel Automatic Grouping Algorithm for Feature Selection publication-title: CCF Chinese Conference on Computer Vision contributor: fullname: Fang – year: 2012 ident: bib0033 article-title: Elements of Information Theory contributor: fullname: Thomas – volume: 403 start-page: 503 year: 2000 end-page: 511 ident: bib0049 article-title: Distinct types of diffuse large b-cell lymphoma identified by gene expression profiling publication-title: Nature contributor: fullname: Yu – volume: 22 start-page: 79 year: 1951 end-page: 86 ident: bib0035 article-title: On information and sufficiency publication-title: Ann. Math. Stat. contributor: fullname: Leibler – volume: 334 start-page: 1518 year: 2011 end-page: 1524 ident: bib0052 article-title: Detecting novel associations in large data sets publication-title: Science contributor: fullname: Sabeti – volume: 27 start-page: 1226 year: 2005 end-page: 1238 ident: bib0020 article-title: Feature selection based on mutual information: Criteria of max-dependency, max-relevance, and min-redundancy publication-title: IEEE Trans. Pattern Anal. Mach. Intell. contributor: fullname: Ding – volume: 67 start-page: 91 year: 2005 end-page: 108 ident: bib0030 article-title: Sparsity and smoothness via the fused lasso publication-title: J. R. Stat. Soc. Ser. B Stat. Methodol. contributor: fullname: Knight – volume: 27 start-page: 328 year: 2005 end-page: 340 ident: bib0044 article-title: Face recognition using laplacianfaces publication-title: IEEE Trans. Pattern Anal. Mach. Intell. contributor: fullname: Zhang – volume: 5 start-page: 537 year: 1994 end-page: 550 ident: bib0038 article-title: Using mutual information for selecting features in supervised neural net learning publication-title: IEEE Trans. Neural Netw. contributor: fullname: Battiti – volume: 93 start-page: 1408 year: 2013 end-page: 1425 ident: bib0008 article-title: Sparse representation and learning in visual recognition: theory and applications publication-title: Signal Proces. contributor: fullname: Chen – volume: 12 start-page: 687 year: 2000 end-page: 693 ident: bib0019 article-title: Data visualization and feature selection: new algorithms for nongaussian data publication-title: Adv. Neural Inf. Process. Syst. contributor: fullname: Moody – volume: 24 start-page: 175 year: 2014 end-page: 186 ident: bib0036 article-title: A review of feature selection methods based on mutual information publication-title: Neural Comput. Appl. contributor: fullname: Estvez – volume: 98 start-page: 1031 year: 2010 end-page: 1044 ident: bib0007 article-title: Sparse representation for computer vision and pattern recognition publication-title: Proc. IEEE contributor: fullname: Yan – volume: 5 start-page: 1531 year: 2003 end-page: 1555 ident: bib0039 article-title: Fast binary feature selection with conditional mutual information publication-title: J. Mach. Learn. Res. contributor: fullname: Fleuret – year: 2010 ident: bib0009 article-title: Sparse and Redundant Representations: From Theory to Applications in Signal and Image Processing contributor: fullname: Elad – volume: PP start-page: 1 year: 2016 end-page: 18 ident: bib0026 article-title: Feature selection based on structured sparsity: a comprehensive study publication-title: IEEE Trans. Neural Netw. Learn. Syst. contributor: fullname: Tan – volume: 41 start-page: 452 year: 1987 end-page: 458 ident: bib0043 article-title: Jensen’s inequality publication-title: Math. Notes contributor: fullname: Guseinov – volume: 73 start-page: 267 year: 2011 end-page: 288 ident: bib0040 article-title: Regression shrinkage and selection via the lasso: a retrospective publication-title: J. R. Stat. Soc. contributor: fullname: Tibshirani – volume: 32 start-page: 407 year: 2004 end-page: 451 ident: bib0042 article-title: Least angle regression publication-title: Ann. Stat. contributor: fullname: Tibshirani – start-page: 3819 year: 2015 end-page: 3827 ident: bib0045 article-title: Two birds, one stone: Jointly learning binary code for large-scale face image retrieval and attributes prediction publication-title: Proceedings of the IEEE International Conference on Computer Vision contributor: fullname: Jiang – start-page: 212 year: 1992 end-page: 217 ident: bib0037 article-title: Feature selection and feature extraction for text categorization publication-title: The Workshop on Speech & Natural Language contributor: fullname: Lewis – start-page: 37 year: 2014 end-page: 64 ident: bib0002 article-title: Feature selection for classification: A review publication-title: Data Classification contributor: fullname: Aggarwal – volume: 83 start-page: 164 year: 2009 end-page: 177 ident: bib0011 article-title: A regularized framework for feature selection in face detection and authentication publication-title: Int. J. Comput. Vis. contributor: fullname: Verri – year: 2007 ident: bib0028 article-title: A Regularized Approach to Feature Selection for Face Detection contributor: fullname: Verri – year: 1959 ident: bib0034 article-title: Information Theory and Statistics contributor: fullname: Kullback – volume: 1 start-page: 142 year: 1998 end-page: 143 ident: bib0005 article-title: Pattern classification publication-title: Pattern Anal. Appl. contributor: fullname: Guyon – volume: 2009 start-page: 403689 year: 2009 ident: bib0014 article-title: Sparse representation for classification of tumors using gene expression data publication-title: J. Biomed. Biotechnol. contributor: fullname: Xiyi Hang – volume: 6 start-page: e28210 year: 2011 ident: bib0001 article-title: The influence of feature selection methods on accuracy, stability and interpretability of molecular signatures publication-title: Plos One contributor: fullname: Jean-Philippe – volume: 23 start-page: 2507 year: 2007 end-page: 2517 ident: bib0003 article-title: Wld: review of feature selection techniques in bioinformatics publication-title: Bioinformatics contributor: fullname: Aga – volume: 36 start-page: 587 year: 2008 end-page: 613 ident: bib0031 article-title: Asymptotic properties of bridge estimators in sparse high-dimensional regression models publication-title: Ann. Stat. contributor: fullname: Ma – volume: 53 start-page: 46 year: 2016 end-page: 58 ident: bib0050 article-title: Can high-order dependencies improve mutual information based feature selection? publication-title: Pattern Recognit. contributor: fullname: Bailey – volume: 42 start-page: 838 year: 2012 end-page: 849 ident: bib0032 article-title: Automatic image annotation and retrieval using group sparsity publication-title: IEEE Trans. Syst. Man. Cybern. Part B Cybern. contributor: fullname: Metaxas – volume: 6 start-page: 23 year: 2007 ident: bib0048 article-title: Exploiting nonlinear recurrence and fractal scaling properties for voice disorder detection publication-title: BioMed. Eng. OnLine contributor: fullname: Moroz – volume: 10 start-page: 10635 year: 2014 end-page: 10642 ident: bib0015 article-title: Sparse representation for feature selection in face demographic classification publication-title: J. Comput. Inf. Syst. contributor: fullname: Fang – volume: 67 start-page: 768 year: 2005 ident: bib0018 article-title: Regularization and variable selection via the elastic net publication-title: J. R. Stat. Soc. contributor: fullname: Hastie – volume: 13 start-page: 27 year: 2012 end-page: 66 ident: bib0021 article-title: Conditional likelihood maximisation: a unifying framework for information theoretic feature selection publication-title: J. Mach. Learn. Res. contributor: fullname: LujÃ!‘n – volume: 101 start-page: 1418 year: 2006 end-page: 1429 ident: bib0029 article-title: The adaptive lasso and its oracle properties publication-title: J. Am. Stat. Assoc. contributor: fullname: Zou – start-page: 78 year: 2004 ident: bib0025 article-title: Feature selection, L 1 vs. L 2 regularization, and rotational invariance publication-title: Proceedings of the twenty-first international conference on Machine learning contributor: fullname: Ng – volume: Vol. 7950 start-page: 197 year: 2013 ident: bib0046 article-title: Evaluation of features for leaf discrimination publication-title: Proceedings of the 10th International Conference on Image Analysis and Recognition, ICIAR, Aveiro, Portugal contributor: fullname: da Silva – volume: 17 start-page: 1433 year: 2006 end-page: 1452 ident: bib0022 article-title: Compressive sampling publication-title: Marta Sanz Sol contributor: fullname: Cands – volume: 3 start-page: 1157 year: 2003 end-page: 1182 ident: bib0004 article-title: An introduction to variable and feature selection publication-title: J. Mach. Learn. Res. contributor: fullname: Elisseeff – volume: 28 start-page: 2037 year: 2006 end-page: 2041 ident: bib0053 article-title: Face description with local binary patterns: application to face recognition publication-title: Pattern Anal. Mach. Intell. IEEE Trans. contributor: fullname: Pietikainen – start-page: 1324 year: 2011 end-page: 1329 ident: bib0027 article-title: Feature selection via joint embedding learning and sparse regression publication-title: Proceedings of the International Joint Conference on Artificial Intelligence, IJCAI, Barcelona, Catalonia, Spain, July contributor: fullname: Wu – volume: 51 start-page: 34 year: 2009 end-page: 81 ident: bib0024 article-title: From sparse solutions of systems of equations to sparse modeling of signals and images publication-title: Soc. Ind. Appl. Math Rev contributor: fullname: Elad – volume: 59 start-page: 797 year: 2007 end-page: 829 ident: bib0023 article-title: For most large underdetermined systems of linear equations the minimal l1norm solution is also the sparsest solution publication-title: Commun. Pure Appl. Math. contributor: fullname: Donoho – volume: Vol. 2 start-page: 1813 year: 2010 end-page: 1821 ident: bib0013 article-title: Efficient and robust feature selection via joint ℓ2,1-norms minimization publication-title: Proceedings of the 23rd International Conference on Neural Information Processing Systems contributor: fullname: Ding – start-page: 224 year: 2015 end-page: 233 ident: bib0012 article-title: Multi-instance feature learning based on sparse representation for facial expression recognition publication-title: Proceedings of the International Conference on Multimedia Modeling contributor: fullname: Chang – start-page: 3819 year: 2015 ident: 10.1016/j.neucom.2018.09.017_bib0045 article-title: Two birds, one stone: Jointly learning binary code for large-scale face image retrieval and attributes prediction contributor: fullname: Li – start-page: 1324 year: 2011 ident: 10.1016/j.neucom.2018.09.017_bib0027 article-title: Feature selection via joint embedding learning and sparse regression contributor: fullname: Hou – volume: 27 start-page: 328 issue: 3 year: 2005 ident: 10.1016/j.neucom.2018.09.017_bib0044 article-title: Face recognition using laplacianfaces publication-title: IEEE Trans. Pattern Anal. Mach. Intell. doi: 10.1109/TPAMI.2005.55 contributor: fullname: He – volume: 23 start-page: 2507 issue: 19 year: 2007 ident: 10.1016/j.neucom.2018.09.017_bib0003 article-title: Wld: review of feature selection techniques in bioinformatics publication-title: Bioinformatics doi: 10.1093/bioinformatics/btm344 contributor: fullname: Saeys – volume: 12 start-page: 687 year: 2000 ident: 10.1016/j.neucom.2018.09.017_bib0019 article-title: Data visualization and feature selection: new algorithms for nongaussian data publication-title: Adv. Neural Inf. Process. Syst. contributor: fullname: Yang – volume: 13 start-page: 27 issue: 1 year: 2012 ident: 10.1016/j.neucom.2018.09.017_bib0021 article-title: Conditional likelihood maximisation: a unifying framework for information theoretic feature selection publication-title: J. Mach. Learn. Res. contributor: fullname: Brown – start-page: 78 year: 2004 ident: 10.1016/j.neucom.2018.09.017_bib0025 article-title: Feature selection, L 1 vs. L 2 regularization, and rotational invariance contributor: fullname: Ng – volume: 2009 start-page: 403689 issue: 1 year: 2009 ident: 10.1016/j.neucom.2018.09.017_bib0014 article-title: Sparse representation for classification of tumors using gene expression data publication-title: J. Biomed. Biotechnol. contributor: fullname: Xiyi Hang – volume: 73 start-page: 267 issue: 3 year: 2011 ident: 10.1016/j.neucom.2018.09.017_bib0040 article-title: Regression shrinkage and selection via the lasso: a retrospective publication-title: J. R. Stat. Soc. doi: 10.1111/j.1467-9868.2011.00771.x contributor: fullname: Tibshirani – ident: 10.1016/j.neucom.2018.09.017_bib0047 – volume: 6 start-page: 23 issue: 1 year: 2007 ident: 10.1016/j.neucom.2018.09.017_bib0048 article-title: Exploiting nonlinear recurrence and fractal scaling properties for voice disorder detection publication-title: BioMed. Eng. OnLine doi: 10.1186/1475-925X-6-23 contributor: fullname: Little – volume: 1 start-page: 142 issue: 2 year: 1998 ident: 10.1016/j.neucom.2018.09.017_bib0005 article-title: Pattern classification publication-title: Pattern Anal. Appl. doi: 10.1007/BF01237942 contributor: fullname: Guyon – volume: 26 start-page: 1388 issue: 7 year: 2015 ident: 10.1016/j.neucom.2018.09.017_bib0041 article-title: Frel: a stable feature selection algorithm publication-title: IEEE Trans. Neural Netw. Learn. Syst. doi: 10.1109/TNNLS.2014.2341627 contributor: fullname: Li – volume: 67 start-page: 768 issue: 2 year: 2005 ident: 10.1016/j.neucom.2018.09.017_bib0018 article-title: Regularization and variable selection via the elastic net publication-title: J. R. Stat. Soc. doi: 10.1111/j.1467-9868.2005.00527.x contributor: fullname: Zou – volume: PP start-page: 1 issue: 99 year: 2016 ident: 10.1016/j.neucom.2018.09.017_bib0026 article-title: Feature selection based on structured sparsity: a comprehensive study publication-title: IEEE Trans. Neural Netw. Learn. Syst. contributor: fullname: Gui – start-page: 224 year: 2015 ident: 10.1016/j.neucom.2018.09.017_bib0012 article-title: Multi-instance feature learning based on sparse representation for facial expression recognition contributor: fullname: Fang – volume: 10 start-page: 10635 issue: 24 year: 2014 ident: 10.1016/j.neucom.2018.09.017_bib0015 article-title: Sparse representation for feature selection in face demographic classification publication-title: J. Comput. Inf. Syst. contributor: fullname: Yu – volume: 93 start-page: 1408 issue: 6 year: 2013 ident: 10.1016/j.neucom.2018.09.017_bib0008 article-title: Sparse representation and learning in visual recognition: theory and applications publication-title: Signal Proces. doi: 10.1016/j.sigpro.2012.09.011 contributor: fullname: Cheng – volume: 51 start-page: 34 issue: 1 year: 2009 ident: 10.1016/j.neucom.2018.09.017_bib0024 article-title: From sparse solutions of systems of equations to sparse modeling of signals and images publication-title: Soc. Ind. Appl. Math Rev contributor: fullname: Bruckstein – year: 2007 ident: 10.1016/j.neucom.2018.09.017_bib0028 contributor: fullname: Destrero – volume: 334 start-page: 1518 issue: 6062 year: 2011 ident: 10.1016/j.neucom.2018.09.017_bib0052 article-title: Detecting novel associations in large data sets publication-title: Science doi: 10.1126/science.1205438 contributor: fullname: Reshef – volume: 22 start-page: 79 issue: 1 year: 1951 ident: 10.1016/j.neucom.2018.09.017_bib0035 article-title: On information and sufficiency publication-title: Ann. Math. Stat. doi: 10.1214/aoms/1177729694 contributor: fullname: Kullback – ident: 10.1016/j.neucom.2018.09.017_bib0006 – volume: Vol. 7950 start-page: 197 year: 2013 ident: 10.1016/j.neucom.2018.09.017_bib0046 article-title: Evaluation of features for leaf discrimination contributor: fullname: Silva – volume: 59 start-page: 797 issue: 6 year: 2007 ident: 10.1016/j.neucom.2018.09.017_bib0023 article-title: For most large underdetermined systems of linear equations the minimal l1norm solution is also the sparsest solution publication-title: Commun. Pure Appl. Math. doi: 10.1002/cpa.20132 contributor: fullname: Donoho – volume: 52 start-page: 1289 issue: 4 year: 2006 ident: 10.1016/j.neucom.2018.09.017_bib0017 article-title: Compressed sensing publication-title: IEEE Trans. Inf. Theory doi: 10.1109/TIT.2006.871582 contributor: fullname: Donoho – year: 1959 ident: 10.1016/j.neucom.2018.09.017_bib0034 contributor: fullname: Kullback – start-page: 212 year: 1992 ident: 10.1016/j.neucom.2018.09.017_bib0037 article-title: Feature selection and feature extraction for text categorization contributor: fullname: Lewis – volume: 403 start-page: 503 issue: 6769 year: 2000 ident: 10.1016/j.neucom.2018.09.017_bib0049 article-title: Distinct types of diffuse large b-cell lymphoma identified by gene expression profiling publication-title: Nature doi: 10.1038/35000501 contributor: fullname: Alizadeh – year: 2010 ident: 10.1016/j.neucom.2018.09.017_bib0009 contributor: fullname: Elad – volume: Vol. 2 start-page: 1813 year: 2010 ident: 10.1016/j.neucom.2018.09.017_bib0013 article-title: Efficient and robust feature selection via joint ℓ2,1-norms minimization contributor: fullname: Nie – volume: 32 start-page: 407 issue: 2 year: 2004 ident: 10.1016/j.neucom.2018.09.017_bib0042 article-title: Least angle regression publication-title: Ann. Stat. doi: 10.1214/009053604000000067 contributor: fullname: Efron – year: 2012 ident: 10.1016/j.neucom.2018.09.017_bib0033 contributor: fullname: Cover – volume: 5 start-page: 1531 issue: 3 year: 2003 ident: 10.1016/j.neucom.2018.09.017_bib0039 article-title: Fast binary feature selection with conditional mutual information publication-title: J. Mach. Learn. Res. contributor: fullname: Fleuret – volume: 42 start-page: 838 issue: 3 year: 2012 ident: 10.1016/j.neucom.2018.09.017_bib0032 article-title: Automatic image annotation and retrieval using group sparsity publication-title: IEEE Trans. Syst. Man. Cybern. Part B Cybern. doi: 10.1109/TSMCB.2011.2179533 contributor: fullname: Zhang – volume: 27 start-page: 1226 issue: 8 year: 2005 ident: 10.1016/j.neucom.2018.09.017_bib0020 article-title: Feature selection based on mutual information: Criteria of max-dependency, max-relevance, and min-redundancy publication-title: IEEE Trans. Pattern Anal. Mach. Intell. doi: 10.1109/TPAMI.2005.159 contributor: fullname: Peng – volume: 24 start-page: 175 issue: 1 year: 2014 ident: 10.1016/j.neucom.2018.09.017_bib0036 article-title: A review of feature selection methods based on mutual information publication-title: Neural Comput. Appl. doi: 10.1007/s00521-013-1368-0 contributor: fullname: Vergara – volume: 17 start-page: 1433 issue: 2 year: 2006 ident: 10.1016/j.neucom.2018.09.017_bib0022 article-title: Compressive sampling publication-title: Marta Sanz Sol contributor: fullname: Cands – start-page: 37 year: 2014 ident: 10.1016/j.neucom.2018.09.017_bib0002 article-title: Feature selection for classification: A review contributor: fullname: Aggarwal – volume: 83 start-page: 164 issue: 2 year: 2009 ident: 10.1016/j.neucom.2018.09.017_bib0011 article-title: A regularized framework for feature selection in face detection and authentication publication-title: Int. J. Comput. Vis. doi: 10.1007/s11263-008-0180-2 contributor: fullname: Destrero – volume: 6 start-page: e28210 issue: 12 year: 2011 ident: 10.1016/j.neucom.2018.09.017_bib0001 article-title: The influence of feature selection methods on accuracy, stability and interpretability of molecular signatures publication-title: Plos One doi: 10.1371/journal.pone.0028210 contributor: fullname: Anne-Claire – volume: 3 start-page: 1157 issue: 6 year: 2003 ident: 10.1016/j.neucom.2018.09.017_bib0004 article-title: An introduction to variable and feature selection publication-title: J. Mach. Learn. Res. contributor: fullname: Guyon – volume: 53 start-page: 46 year: 2016 ident: 10.1016/j.neucom.2018.09.017_bib0050 article-title: Can high-order dependencies improve mutual information based feature selection? publication-title: Pattern Recognit. doi: 10.1016/j.patcog.2015.11.007 contributor: fullname: Vinh – volume: 101 start-page: 1418 issue: 476 year: 2006 ident: 10.1016/j.neucom.2018.09.017_bib0029 article-title: The adaptive lasso and its oracle properties publication-title: J. Am. Stat. Assoc. doi: 10.1198/016214506000000735 contributor: fullname: Zou – volume: 41 start-page: 452 issue: 6 year: 1987 ident: 10.1016/j.neucom.2018.09.017_bib0043 article-title: Jensen’s inequality publication-title: Math. Notes doi: 10.1007/BF01158386 contributor: fullname: Guseinov – volume: 36 start-page: 587 issue: 2 year: 2008 ident: 10.1016/j.neucom.2018.09.017_bib0031 article-title: Asymptotic properties of bridge estimators in sparse high-dimensional regression models publication-title: Ann. Stat. doi: 10.1214/009053607000000875 contributor: fullname: Huang – volume: 5 start-page: 537 issue: 4 year: 1994 ident: 10.1016/j.neucom.2018.09.017_bib0038 article-title: Using mutual information for selecting features in supervised neural net learning publication-title: IEEE Trans. Neural Netw. doi: 10.1109/72.298224 contributor: fullname: Battiti – volume: 67 start-page: 91 issue: 1 year: 2005 ident: 10.1016/j.neucom.2018.09.017_bib0030 article-title: Sparsity and smoothness via the fused lasso publication-title: J. R. Stat. Soc. Ser. B Stat. Methodol. doi: 10.1111/j.1467-9868.2005.00490.x contributor: fullname: Tibshirani – volume: 18 start-page: 188 issue: 1 year: 2009 ident: 10.1016/j.neucom.2018.09.017_bib0010 article-title: A sparsity-enforcing method for learning face features publication-title: IEEE Trans. Image Process. doi: 10.1109/TIP.2008.2007610 contributor: fullname: Destrero – volume: 28 start-page: 2037 issue: 12 year: 2006 ident: 10.1016/j.neucom.2018.09.017_bib0053 article-title: Face description with local binary patterns: application to face recognition publication-title: Pattern Anal. Mach. Intell. IEEE Trans. doi: 10.1109/TPAMI.2006.244 contributor: fullname: Ahonen – start-page: 592 year: 2017 ident: 10.1016/j.neucom.2018.09.017_bib0016 article-title: A Novel Automatic Grouping Algorithm for Feature Selection contributor: fullname: Yuan – start-page: 68 year: 2006 ident: 10.1016/j.neucom.2018.09.017_bib0051 article-title: Conditional infomax learning: an integrated framework for feature extraction and fusion contributor: fullname: Lin – volume: 98 start-page: 1031 issue: 6 year: 2010 ident: 10.1016/j.neucom.2018.09.017_bib0007 article-title: Sparse representation for computer vision and pattern recognition publication-title: Proc. IEEE doi: 10.1109/JPROC.2010.2044470 contributor: fullname: Wright |
SSID | ssj0017129 |
Score | 2.271016 |
Snippet | Feature selection is used in many applications in machine learning and bioinformatics. As a popular approach, feature selection can be implemented in the... |
SourceID | crossref elsevier |
SourceType | Aggregation Database Publisher |
StartPage | 237 |
SubjectTerms | Feature selection Information theory l1 regularization Mutual information |
Title | On the role of sparsity in feature selection and an innovative method LRMI |
URI | https://dx.doi.org/10.1016/j.neucom.2018.09.017 |
Volume | 321 |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV07T8MwED6VdmHhjSiPygOraeK4TjJWFVVbaJGASt0s23GkMqQVtIz8ds5OgkBCDAwZ8rCU3MV33yWf7wO4zsJUax4pGuZpTHmiA5xSIqIqFgHLUmOsbzw_nYnRnE8WvUUDBvVaGEerrGJ_GdN9tK6OdCtrdtfLZfcpSBlWUaHL8VjDJGIHWpiOOG9Cqz--G82-fibEIStb7rEedQPqFXSe5lXYraONYB5MfMNTr1z2S4b6lnWGB7BXwUXSL-_oEBq2OIL9WoqBVDPzGCYPBUEoRxxZkKxygnHCsy3IsiC59c07yZuXvEE_EFVkuOG5UhH13ZJSSJrcP07HJzAf3j4PRrSSSaAG64ENZYlAN0TaIPbJEmU5IiRj8XlY0mOWx9q1IDMZ1kUqirnAgtgkGKfQL6HK3VfhU2gWq8KeAUFskCnBApOrnBujNXMAxgoVZ8pd3QZam0auy24YsqaJvcjSlNKZUgapRFO2Ia7tJ394VWLA_nPk-b9HXsCu23OUkzC4hObmdWuvEDhsdAd2bj7CTvV6fALSwcHl |
link.rule.ids | 315,786,790,4521,24144,27957,27958,45620,45714 |
linkProvider | Elsevier |
linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV07T8MwED6VdoCFN6I8PbBaTRzHScaqouobCVqpm-U4jlSGtIKW38_ZSRBIiIEhSxxLyXfx3XfJ5zuAh8xP0pQHivp5ElEepx4uKRFQFQmPZYnWxhWen87EYMFHy3DZgF69F8bKKivfX_p0562rM50Kzc5mteq8eAnDLMq3MR5zmFjsQYuHkc-b0OoOx4PZ18-EyGdlyT0WUjuh3kHnZF6F2VnZCMbB2BU8dZ3LfolQ36JO_xgOK7pIuuUdnUDDFKdwVLdiINXKPIPRU0GQyhErFiTrnKCfcGoLsipIblzxTvLuWt6gHYgqMjxwrOyI-mFI2UiaTJ6nw3NY9B_nvQGt2iRQjfnAlrJYoBmCVCP3yWJlODIkbfB5WBwyw6PUliDTGeZFKoi4wIRYx-in0C6-yu1X4QtoFuvCXAJBbpApwTydq5xrnabMEhgjVJQpe3UbaA2N3JTVMGQtE3uVJZTSQim9RCKUbYhq_OQPq0p02H_OvPr3zHvYH8ynEzkZzsbXcGBHrPzE926guX3bmVskEdv0rnpJPgGEccPX |
openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=On+the+role+of+sparsity+in+feature+selection+and+an+innovative+method+LRMI&rft.jtitle=Neurocomputing+%28Amsterdam%29&rft.au=Fang%2C+Yuchun&rft.au=Yuan%2C+Qiulong&rft.au=Zhang%2C+Zhaoxiang&rft.date=2018-12-10&rft.pub=Elsevier+B.V&rft.issn=0925-2312&rft.eissn=1872-8286&rft.volume=321&rft.spage=237&rft.epage=250&rft_id=info:doi/10.1016%2Fj.neucom.2018.09.017&rft.externalDocID=S0925231218310786 |
thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0925-2312&client=summon |
thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0925-2312&client=summon |
thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0925-2312&client=summon |