An introduction to variable and feature selection

By Isabelle M Guyon, Andre´ Elisseeﬀ. 2003

1,079 citations recorded by the source corpus.

Variable and feature selection have become the focus of much research in areas of application for which datasets with tens or hundreds of thousands of variables are available. These areas include text processing of internet documents, gene expression array analysis, and combinatorial chemistry. The objective of variable selection is three-fold: improving the prediction performance of the predictors, providing faster and more cost-effective predictors, and providing a better understanding of the underlying process that generated the data. The contributions of this special issue cover a wide range of aspects of such problems: providing a better definition of the objective function, feature construction, feature ranking, multivariate feature selection, efficient search methods, and feature validity assessment methods.

Subfields: General Health Professions, Philosophy

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In General Health Professions

In Philosophy

Citations in the corpus

Unmasking Clever Hans predictors and assessing what machines really learn (2019, background): “While feature selection has traditionally explained the model by identifying features relevant for the whole ensemble of training data [12] or some class prototype [13–16], it is often necessary, especially for nonlinear models, to focus the explanation on the predictions of individual examples (see”
How to Explain Individual Classification Decisions (2009, methodology): “for all instances of the training set, for example feature selection methods (including Automatic Relevance Determination) ﬁnd out which inputs are salient for a good generalization (see for a review Guyon and Elisseeﬀ, 2003). While this can give a coarse impression about the global usefulness of ea”
PRoNTo: Pattern Recognition for Neuroimaging Toolbox (2013, background): “…and temporal compression, which are particularly useful in a within-subject context (Mourao-Miranda et al. 2006), feature selection could bring valuable advantages such as reducing memory storage requirements (Formisiano et al. 2008) and possibly improving model accuracy (Guyon and Elisseeff 2007).”
Content-based image retrieval: approaches and trends of the new age (2005, neutral): “A survey and performance comparison of some recent algorithms on the topic can be found in [34].”
Using support vector machines with a novel hybrid feature selection method for diagnosis of erythemato-squamous diseases (2011, methodology): “This paper presents a novel hybrid feature selection method named improved F-score and sequential forward search (IFSFS), where the improved F-score plays the role of filter, and SFS plays the role of wrapper (Guyon, 2003; Nakariyakul & Casasent, 2007).”
X3D: Expanding Architectures for Efficient Video Recognition (2020, methodology): “In the latter, progressive feature selection methods [21, 34] start with either a set of minimum features and aim to find relevant features to improve in a greedy fashion by including (forward selection) a single feature in each step, or start with a full set of features and aim to find irrelevant o”
Feature Selection with the Boruta Package (2010, background): “Moreover, one cannot use filtering methods, because the lack of direct correlation between a given feature and the decision is not a proof that this feature is not important in conjunction with the other features (Guyon and Elisseeff 2003).”
Applications of machine learning in animal behaviour studies (2017, neutral): “These techniques come in three flavours: filter, wrapper and embedded methods (Guyon & Elisseeff, 2003).”
SISSO: A compressed-sensing method for identifying the best low-dimensional descriptor in an immensity of offered candidates (2017, neutral): “All quantities that are hypothesized to be relevant for describing the target property (the so called primary features [7, 31]) are used as starting point for the construction of the space [37, 38].”
Cell Painting, a high-content image-based assay for morphological profiling using multiplexed fluorescent dyes (2016, methodology): “48| Select the FlagImages module, which is used to label images with a metadata tag of 0 or 1, depending on whether features from particular image channels pass or fail the chosen QC criteria, respectively.”
A statistical approach to rule learning (2006, background): “Dispersion-based feature generation can also be viewed as a contribution to the field of feature selection (Guyon and Elisseeff, 2003).”
Unsupervised Feature Selection Using Nonnegative Spectral Analysis (2012, background): “Introduction The dimension of data is often very high in many domains (Jain and Zongker 1997; Guyon and Elisseeff 2003), such as image and video understanding (Wang et al.”
Soli (2016, methodology): “Like many machine learning tasks, extracting salient features from the radar signals is a critical step for building robust gesture recognition pipeline [Guyon and Elisseeff 2003].”
Spectral feature selection for supervised and unsupervised learning (2007, background): “Feature selection is one efiective means to identify relevant features for dimension reduction ( Guyon & Elisseefi, 2003; Liu & Yu, 2005).”