A review of feature selection methods based on mutual information

By Jorge R. Vergara, Pablo A. Estévez. Neural computing & applications (Print), 2013

1,221 citations recorded by the source corpus.

In this work, we present a review of the state of the art of information-theoretic feature selection methods. The concepts of feature relevance, redundance, and complementarity (synergy) are clearly defined, as well as Markov blanket. The problem of optimal feature selection is defined. A unifying theoretical framework is described, which can retrofit successful heuristic criteria, indicating the approximations made by each method. A number of open problems in the field are presented.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/f7bcec54ca019be092ce74379ef840b64ef809aa

How researchers see this paper

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

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

A Survey on Evolutionary Computation Approaches to Feature Selection (2016, background): “ture selection can be seen in [7], [8], [12], and [13], which focus mainly on non-EC-based methods.”
Feature selection in machine learning: A new perspective (2018, background): “feature selection [66-71], they are mainly focus on specific research fields in feature selection.”
A comprehensive survey on recent metaheuristics for feature selection (2022, background): “Vergara & Estévez presented a review for the information-theoretic methods [68].”
Feature selection for text classification: A review (2018, background): “A detailed review on feature selection methods based on mutual information can be found in [95].”
Feature selection for high-dimensional data (2016, background): “Along the same line, Vergara and Estévez [72] presented a review of the state of the art of informationtheoretic feature selection methods.”
Structural Agnostic Modeling: Adversarial Learning of Causal Graphs (2018, background): “It leverages the estimation of mutual information among variables (Bell and Wang, 2000; Brown et al., 2012; Vergara and Estévez, 2014) or uses classification or regression models to support variable selection (Aliferis et al.”
Techniques for Automated Machine Learning (2019, background): “One might integrate domain knowledge or leverage the mutual information from information theory [87] to discern useful features.”