Using Random Forest to Learn Imbalanced Data

By Chao Chen. 2004

1,318 citations recorded by the source corpus.

How researchers see this paper

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

Citations in the corpus

Imbalanced Continual Learning with Partitioning Reservoir Sampling (2020, background): “Consequently, to effectively perform continual learning on multi-label data, two major obstacles should be overcome simultaneously: (i) the infamous catastrophic forgetting problem [49,55,23] and (ii) the long-tailed distribution problem [14,29,66,43], which we jointly address in this work.”
Overview of random forest methodology and practical guidance with emphasis on computational biology and bioinformatics (2012, methodology): “Chen et al. [16] introduce a variant of the RF method based on this idea, the so called ‘Weighted RF’.”
Customer churn prediction using improved balanced random forests (2009, methodology): “In addition, assigning a weight to the minority class may make the method more vulnerable to noise (mislabeled class) (Chen et al., 2004).”
Learning Deep Representation for Imbalanced Classification (2016, background): “, [9, 41]), and boosting [41] offers an easy way to embed the costs by updating example weights.”
Churn in Social Networks (2010, background): “Further, [8] concludes that advanced sampling techniques do not provide significant improvement and that advanced classification techniques, such as cost-sensitive learners [67] like weighted random forests [10] and boosting [19], should always be compared to logistic regression.”
Detecting click fraud in online advertising: a data mining approach (2014, methodology): “For random forests, both approaches were shown to be on par in terms of performance (Chen et al., 2004).”
Modeling Species Distribution and Change Using Random Forest (2011, background): “One issue with Random Forest arises within classification problems when classes in categorical response variables are imbalanced (Chen et al. 2004).”
Comparing Random Forest with Logistic Regression for Predicting Class-Imbalanced Civil War Onset Data (2016, methodology): “Such a procedure allows the algorithm to weight the relative importance of each class according to such a ratio, and to generate correct predictions of rare events even in extremely imbalanced data (Chen et al. 2004; Weidmann 2008).”
Tree Species Classification in Boreal Forests With Hyperspectral Data (2013, background): “RF in the presence of unbalanced data sets has the tendency to privilege majority classes [40], while GML with a low number of training samples is not able to parameterize the”
Robust Nucleus/Cell Detection and Segmentation in Digital Pathology and Microscopy Images: A Comprehensive Review (2016, methodology): “Recently, Irshad et al. [21] have reported a survey on the methods for nucleus detection, segmentation, feature extraction, and classification on hematoxylin and eosin (H&E) Description CAD computer-aided diagnosis SIFT scale-invariant feature transform MAP maximum a posteriori WSI whole-slide imagi”
A Survey of Predictive Modeling on Imbalanced Domains (2016, methodology): “Namely, it was combined with boosting [Seiffert et al. 2010] and bagging [Chang et al. 2003; Tao et al. 2006; Wang and Yao 2009; Wallace et al. 2011] and was applied to both classes in random forests in a method named Balanced Random Forest (BRF) [Chen et al. 2004].”
APATE: A novel approach for automated credit card transaction fraud detection using network-based extensions (2015, methodology): “…random forests we will use the sub-sampling capabilities of random forests, with each tree constructed using all fraudulent transactions and a randomly selected subset of the non-fraudulent ones such that they account for two times the number of fraudulent ones, as explained in Chen et al. (2004).”
On the use of MapReduce for imbalanced big data using Random Forest (2014, methodology): “Please cite this article in press as: S. del Río et al., On the use of MapReduce for imbalanced big data using Random Forest, Inform.”
Handling class imbalance in customer churn prediction (2009, result): “This explains why some studies have shown simple over-sampling to be ineffective at improving recognition of the minority class (Ling & Li, 1998; Drummond & Holte, 2003) and why under-sampling seems to have an edge over over-sampling (Chen et al., 2004).”
Mean Birds: Detecting Aggression and Bullying on Twitter (2017, background): “Based on [9], and similar to almost all classi-”
Class Rectification Hard Mining for Imbalanced Deep Learning (2017, background): “(2) Cost-sensitive learning : Assigning higher misclassiﬁcation costs to the minority classes as compared to the majority classes [44, 49, 5, 51, 43], or regularising the cross-entropy loss to cope with the imbalanced positive and negative class distribution [40].”