Enriched random forests

By D. Amaratunga, Javier Cabrera, Yung-Seop Lee. 2008

232 citations recorded by the source corpus.

Although the random forest classification procedure works well in datasets with many features, when the number of features is huge and the percentage of truly informative features is small, such as with DNA microarray data, its performance tends to decline significantly. In such instances, the procedure can be improved by reducing the contribution of trees whose nodes are populated by non-informative features. To some extent, this can be achieved by prefiltering, but we propose a novel, yet simple, adjustment that has demonstrably superior performance: choose the eligible subsets at each node by weighted random sampling instead of simple random sampling, with the weights tilted in favor of the informative features. This results in an 'enriched random forest'. We illustrate the superior performance of this procedure in several actual microarray datasets.

Subfields: Computer Vision and Pattern Recognition, Molecular Biology, Statistics and Probability

How researchers see this paper

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

In Computer Vision and Pattern Recognition

In Molecular Biology

In Statistics and Probability

Citations in the corpus

A Review of Ensemble Methods in Bioinformatics (2010, methodology): “[96] designed a random forests variant called enriched random forests which weigh the importance of features when selecting splitting nodes.”
Fast and Flexible Monotonic Functions with Ensembles of Lattices (2016, background): “Sampling the features to increase the likelihood of selecting combinations of features that better correlate with the label can produce better random forests [13].”
Random forests: from early developments to recent advancements (2014, methodology): “Used weighted voting instead of majority voting Tsymbal et al. (2006) Dynamic integration techniques Replaced majority voting with more sophisticated dynamic integration techniques: DS, DV, and DVS Amaratunga et al. (2008) Weighted random sampling Improved the declining performance when the number o”
A Brief Review of Random Forests for Water Scientists and Practitioners and Their Recent History in Water Resources (2019, methodology): “It corresponds to a lower parametric version of random forests Enriched random forests [104] Weighted random selection of predictor variables as candidates for splitting.”
Overview of random forest methodology and practical guidance with emphasis on computational biology and bioinformatics (2012, background): “They have been considered in tens of bioinformatics papers [81, 62, 80, 2, 44] and biomedical applications [12, 46, 10, 56, 78, 15, 70, 47] including genome-wide studies [63, 79, 31, 64, 51].”
Case-Specific Random Forests (2016, methodology): “…random forests (De’ath 2010; Segal and Xiao 2011), quantile regression forests (Mein-shausen 2006), enriched random forests for microarray analysis (Amaratunga, Cabrera, and Lee 2008), random survival forests (Ishwaran et al. 2008), the R package pathwayRF for metabolic pathway analysis (Pang et…”
Random Forests for Genomic Data Analysis (2012, background): “It has achieved competitive predictionresultsona benchmarkexperimentinvolvingtenmicroarray datasets [25].”