Variable Selection Using Random Forests

By Marco Sandri, Paola Zuccolotto. 2006

879 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Control and Systems Engineering

View Paper: https://www.semanticscholar.org/paper/a8e3560f5584c37aaf5a2fdb3b82c0f2243e9508

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 Control and Systems Engineering

Citations in the corpus

Random Forests for Genomic Data Analysis (2012, methodology): “Genuer et al. [18] developed another heuristic strategy of variable selection using RF.”
A review of feature selection methods with applications (2015, background): “5, random forest [24], but also other algorithms (e.”
An exploratory study of the pull-based software development model (2014, background): “The selected set of features should then be enough to predict the classification outcome with reasonable accuracy, and therefore can be described as important [12].”
Explainable AI for Trees: From Local Explanations to Global Understanding (2019, methodology): “This is particularly important since tree-based models are often used in practice for feature selection [22, 50].”
Grouped variable importance with random forests and application to multiple functional data analysis (2014, methodology): “In the context of random forests (Breiman; 2001), it has been shown that the permutation importance measure introduced by Breiman, is an eﬃcient tool for selecting variables (D´ıaz-Uriarte and Alvarez de Andr´es; 2006; Genuer et al.; 2010; Gregorutti et al.; 2014).”
Overview of random forest methodology and practical guidance with emphasis on computational biology and bioinformatics (2012, methodology): “An alternative variable selection approach based on a nested collection of random forests is described in Genuer et al. [29].”
Land-use/cover classification in a heterogeneous coastal landscape using RapidEye imagery: evaluating the performance of random forest and support vector machines classifiers (2014, background): “To ensure lower similarity (i.e. diversity) between individual trees and thus a low bias, no pruning is performed and all trees in the forest are maximal-grown trees (Genuer, Poggi, and Tuleau-Malot 2010; Lin et al. 2010; Breiman 2001).”
A new variable importance measure for random forests with missing data (2012, background): “The work of Tang et al. (2009), Yang and Gu (2009), Rodenburg et al. (2008), Sandri and Zuccolotto (2006) and Díaz-Uriarte and Alvarez de Andrés (2006) shows that different approaches that are meant to improve prediction accuracy and select the most relevant variables have been developed.”
Correlation and variable importance in random forests (2013, methodology): “More elaborated algorithms based on NRFE has been proposed in the literature as in Genuer et al. (2010). Since we are interested here in the effect of updating the measure importance, we only consider here the original version of NRFE.”
Understanding variable importances in forests of randomized trees (2013, result): “Strobl et al. (2008) later showed that MDA is biased as well, and that it overestimates the importance of correlated variables – although this effect was not confirmed in a later experimental study by Genuer et al. (2010).”
Towards the Deployment of Machine Learning Solutions in Network Traffic Classification: A Systematic Survey (2019, methodology): “embedded into their model design, such as regularized regression models and decision trees based models [30], [31].”
Prediction of adverse cardiac events in emergency department patients with chest pain using machine learning for variable selection (2014, methodology): “The RF approach has been shown to be effective in variable selection [27,28].”
Variable selection using random forests (2010, background): “Finally, let us mention an application currently in progress for fMRI brain activity classification (see Genuer et al. (2010)).”
Hierarchical diversity entropy for the early fault diagnosis of rolling bearing (2021, neutral): “this point, the random forest importance is employed to show the contribution of each components used in HDE [22].”
iForest: Interpreting Random Forests via Visual Analytics (2019, methodology): “, Gini impurity [24]), which determines the feature and its split point for each node in a decision tree [37].”