Discovering informative patterns and data cleaning
Advances in knowledge discovery and data mining
An Algorithm for Finding Best Matches in Logarithmic Expected Time
ACM Transactions on Mathematical Software (TOMS)
Multidimensional binary search trees used for associative searching
Communications of the ACM
Estimating a Kernel Fisher Discriminant in the Presence of Label Noise
ICML '01 Proceedings of the Eighteenth International Conference on Machine Learning
Decontamination of Training Samples for Supervised Pattern Recognition Methods
Proceedings of the Joint IAPR International Workshops on Advances in Pattern Recognition
An introduction to variable and feature selection
The Journal of Machine Learning Research
Computational Statistics & Data Analysis
Machine Learning and Data Mining: Introduction to Principles and Algorithms
Machine Learning and Data Mining: Introduction to Principles and Algorithms
Feature Selection by Transfer Learning with Linear Regularized Models
ECML PKDD '09 Proceedings of the European Conference on Machine Learning and Knowledge Discovery in Databases: Part I
Bayesian analysis of correlated misclassified binary data
Computational Statistics & Data Analysis
Variable selection via combined penalization for high-dimensional data analysis
Computational Statistics & Data Analysis
On the assessment of text corpora
NLDB'09 Proceedings of the 14th international conference on Applications of Natural Language to Information Systems
Nearest neighbor pattern classification
IEEE Transactions on Information Theory
Using mutual information for selecting features in supervised neural net learning
IEEE Transactions on Neural Networks
On selecting interacting features from high-dimensional data
Computational Statistics & Data Analysis
Editorial: Special issue on imprecision in statistical data analysis
Computational Statistics & Data Analysis
| Hi-index | 0.03 |
A way to achieve feature selection for classification problems polluted by label noise is proposed. The performances of traditional feature selection algorithms often decrease sharply when some samples are wrongly labelled. A method based on a probabilistic label noise model combined with a nearest neighbours-based entropy estimator is introduced to robustly evaluate the mutual information, a popular relevance criterion for feature selection. A backward greedy search procedure is used in combination with this criterion to find relevant sets of features. Experiments establish that (i) there is a real need to take a possible label noise into account when selecting features and (ii) the proposed methodology is effectively able to reduce the negative impact of the mislabelled data points on the feature selection process.