Network-based classification using cortical thickness of AD patients
MLMI'11 Proceedings of the Second international conference on Machine learning in medical imaging
Computers in Biology and Medicine
Large-margin feature selection for monotonic classification
Knowledge-Based Systems
Wrappers for web access logs feature selection
Proceedings of the 2nd International Conference on Web Intelligence, Mining and Semantics
A boosting approach for supervised Mahalanobis distance metric learning
Pattern Recognition
Feature weighting by RELIEF based on local hyperplane approximation
PAKDD'12 Proceedings of the 16th Pacific-Asia conference on Advances in Knowledge Discovery and Data Mining - Volume Part II
CLEF'12 Proceedings of the Third international conference on Information Access Evaluation: multilinguality, multimodality, and visual analytics
Minimum-maximum local structure information for feature selection
Pattern Recognition Letters
A local information-based feature-selection algorithm for data regression
Pattern Recognition
Large Margin Subspace Learning for feature selection
Pattern Recognition
Journal of Information Science
Spatial distance join based feature selection
Engineering Applications of Artificial Intelligence
Expert Systems with Applications: An International Journal
Accurate prediction of AD patients using cortical thickness networks
Machine Vision and Applications
PLS-based recursive feature elimination for high-dimensional small sample
Knowledge-Based Systems
Joint Laplacian feature weights learning
Pattern Recognition
Hi-index | 0.15 |
This paper considers feature selection for data classification in the presence of a huge number of irrelevant features. We propose a new feature-selection algorithm that addresses several major issues with prior work, including problems with algorithm implementation, computational complexity, and solution accuracy. The key idea is to decompose an arbitrarily complex nonlinear problem into a set of locally linear ones through local learning, and then learn feature relevance globally within the large margin framework. The proposed algorithm is based on well-established machine learning and numerical analysis techniques, without making any assumptions about the underlying data distribution. It is capable of processing many thousands of features within minutes on a personal computer while maintaining a very high accuracy that is nearly insensitive to a growing number of irrelevant features. Theoretical analyses of the algorithm's sample complexity suggest that the algorithm has a logarithmical sample complexity with respect to the number of features. Experiments on 11 synthetic and real-world data sets demonstrate the viability of our formulation of the feature-selection problem for supervised learning and the effectiveness of our algorithm.