The Balanced Accuracy and Its Posterior Distribution

By Kay Henning Brodersen, Cheng Soon Ong, Klaas E. Stephan, Joachim M. Buhmann. International Conference on Pattern Recognition, 2010

1,729 citations recorded by the source corpus.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/8a097304b1c087d8ad5d25db48bcf6ad91e79ea5

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

Citations in the corpus

Using deep learning to investigate the neuroimaging correlates of psychiatric and neurological disorders: Methods and applications (2017, neutral): “other hand, corresponds to the average accuracy obtained on either class (Brodersen et al., 2010).”
AlignScore: Evaluating Factual Consistency with A Unified Alignment Function (2023, methodology): “(2022), we 1) tune the threshold of metrics on the validation sets, and then compute the balanced accuracy (Brodersen et al., 2010) on the test sets, 2) report the AUC-ROC (Bradley, 1997) of each metric.”
Interpretability is Harder in the Multiclass Setting: Axiomatic Interpretability for Multiclass Additive Models (2018, background): “Balanced accuracy addresses the imbalance of classes in classification tasks (Brodersen et al. 2010): BACC(f) = 1 K ∑K k=1 P(f(x) = k|y = k).”
CatBoost for big data: an interdisciplinary review (2020, neutral): “report the progressive increase in the balanced accuracy [89] scores CatBoost yields as they increase the amount of data they use as input to CatBoost.”
Image Aesthetic Assessment: An experimental survey (2016, background): “uating testing partition with imbalanced distribution [54], [55].”
Metrics for Multi-Class Classification: an Overview (2020, background): “On the other hand, Accuracy treats all instances alike and usually favours the majority class [2].”
Ten quick tips for machine learning in computational biology (2017, background): “1 ), balanced accuracy [33], or F1 score (Eq.”
Decoupling "when to update" from "how to update" (2017, methodology): “Since the amount of male and female subjects in the dataset is not balanced, we use an objective of maximizing the balanced accuracy [9] - the average accuracy obtained on either class.”
A Combined Model- and Learning-Based Framework for Interaction-Aware Maneuver Prediction (2016, methodology): “According to the definition of balanced accuracy in [38], the balanced precision prc and F1 score [39] are defined as”
Malware Detection by Eating a Whole EXE (2017, methodology): “In evaluating the predictive performance of our models, we use Balanced Accuracy (Brodersen et al. 2010) (i.e., accuracy weighted so that benign and malicious samples count evenly) and AUC (Bradley 1997) .”
AdaFair: Cumulative Fairness Adaptive Boosting (2019, methodology): “However in KDD census “the class labels were drawn from the total person income field rather than the adjusted gross income” [2].”
Forecasting the global extent of invasion of the cereal pest Spodoptera frugiperda, the fall armyworm (2018, methodology): “In order to investigate the effect of biased recorder effort and geographic background on environmental suitability, we compared the agreement of the global ensemble projections made using each distribution dataset by using Cohen’s Kappa and balanced accuracy (Brodersen et al. 2010).”
The Matthews correlation coefficient (MCC) is more reliable than balanced accuracy, bookmaker informedness, and markedness in two-class confusion matrix evaluation (2021, methodology): “The scientific community has employed balanced accuracy for a long time, but its benefits over general accuracy were introduced by Brodersen and colleagues [22] in 2010, and reaffirmed by Wei et al.”
Bayesian mixed-effects inference on classification performance in hierarchical data sets (2012, methodology): “Third, our approach permits inference on both the accuracy and the balanced accuracy, a performance measure that avoids bias when working with imbalanced data sets (Brodersen et al., 2010a).”