Advanced search
1 file | 1.08 MB

Fast, accurate, and stable feature selection using neural networks

(2018) NEUROINFORMATICS. 16(2). p.253-268
Author
Organization
Abstract
Multi-voxel pattern analysis often necessitates feature selection due to the high dimensional nature of neuroimaging data. In this context, feature selection techniques serve the dual purpose of potentially increasing classification accuracy and revealing sets of features that best discriminate between classes. However, feature selection techniques in current, widespread use in the literature suffer from a number of deficits, including the need for extended computational time, lack of consistency in selecting features relevant to classification, and only marginal increases in classifier accuracy. In this paper we present a novel method for feature selection based on a single-layer neural network which incorporates cross-validation during feature selection and stability selection through iterative subsampling. Comparing our approach to popular alternative feature selection methods, we find increased classifier accuracy, reduced computational cost and greater consistency with which relevant features are selected. Furthermore, we demonstrate that importance mapping, a technique used to identify voxels relevant to classification, can lead to the selection of irrelevant voxels due to shared activation patterns across categories. Our method, owing to its relatively simple architecture, flexibility and speed, can provide a viable alternative for researchers to identify sets of features that best discriminate classes.
Keywords
MULTIVOXEL PATTERN-ANALYSIS, VENTRAL TEMPORAL CORTEX, SUPPORT VECTOR, MACHINES, MUTUAL INFORMATION, FMRI DATA, VOXEL SELECTION, CLASSIFICATION, STABILITY, REPRESENTATIONS, BIOINFORMATICS, Feature selection, fMRI, MVPA, Machine learning

Downloads

  • (...).pdf
    • full text
    • |
    • UGent only
    • |
    • PDF
    • |
    • 1.08 MB

Citation

Please use this url to cite or link to this publication:

Chicago
Deraeve, James, and William H. Alexander. 2018. “Fast, Accurate, and Stable Feature Selection Using Neural Networks.” Neuroinformatics 16 (2): 253–268.
APA
Deraeve, J., & Alexander, W. H. (2018). Fast, accurate, and stable feature selection using neural networks. NEUROINFORMATICS, 16(2), 253–268.
Vancouver
1.
Deraeve J, Alexander WH. Fast, accurate, and stable feature selection using neural networks. NEUROINFORMATICS. Totowa: Humana Press Inc; 2018;16(2):253–68.
MLA
Deraeve, James, and William H. Alexander. “Fast, Accurate, and Stable Feature Selection Using Neural Networks.” NEUROINFORMATICS 16.2 (2018): 253–268. Print.
@article{8608069,
  abstract     = {Multi-voxel pattern analysis often necessitates feature selection due to the high dimensional nature of neuroimaging data. In this context, feature selection techniques serve the dual purpose of potentially increasing classification accuracy and revealing sets of features that best discriminate between classes. However, feature selection techniques in current, widespread use in the literature suffer from a number of deficits, including the need for extended computational time, lack of consistency in selecting features relevant to classification, and only marginal increases in classifier accuracy. In this paper we present a novel method for feature selection based on a single-layer neural network which incorporates cross-validation during feature selection and stability selection through iterative subsampling. Comparing our approach to popular alternative feature selection methods, we find increased classifier accuracy, reduced computational cost and greater consistency with which relevant features are selected. Furthermore, we demonstrate that importance mapping, a technique used to identify voxels relevant to classification, can lead to the selection of irrelevant voxels due to shared activation patterns across categories. Our method, owing to its relatively simple architecture, flexibility and speed, can provide a viable alternative for researchers to identify sets of features that best discriminate classes.},
  author       = {Deraeve, James and Alexander, William H.},
  issn         = {1539-2791},
  journal      = {NEUROINFORMATICS},
  keywords     = {MULTIVOXEL PATTERN-ANALYSIS,VENTRAL TEMPORAL CORTEX,SUPPORT VECTOR,MACHINES,MUTUAL INFORMATION,FMRI DATA,VOXEL SELECTION,CLASSIFICATION,STABILITY,REPRESENTATIONS,BIOINFORMATICS,Feature selection,fMRI,MVPA,Machine learning},
  language     = {eng},
  number       = {2},
  pages        = {253--268},
  publisher    = {Humana Press Inc},
  title        = {Fast, accurate, and stable feature selection using neural networks},
  url          = {http://dx.doi.org/10.1007/s12021-018-9371-3},
  volume       = {16},
  year         = {2018},
}

Altmetric
View in Altmetric
Web of Science
Times cited: