A systematic analysis of performance measures for classification tasks

By Marina Sokolova, Guy Lapalme. 2009

6,602 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems

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 Information Systems

Citations in the corpus

Recent Advances in Open Set Recognition: A Survey (2018, background): “3) Youden’s index for OSR: As the F-measure is invariant to changes in TN [113], an important factor in OSR performance, Scherreik and Rigling [63] turned to Youden’s index J deﬁned as follows where S = TN/ ( TN + FP ) represents the true negative rate [114].”
Automatic facial expression recognition using features of salient facial patches (2015, background): “The quality of the overall classification is evaluated by calculating the macro-average [60] of preci-”
Multispectral image analysis approach to detect adulteration of beef and pork in raw meats (2015, methodology): “, Sun, D.-W., & Allen, P. (2012). Potential of hyperspectral imaging and pattern recognition for categorization and authentication of red meat. Innovative Food Science & Emerging Technologies, 16, 316–325. doi:10.1016/j.ifset.2012.07.007 Kamruzzaman, M., Sun, D.-W., ElMasry, G., & Allen, P. (2013). ”
COVIDX-Net: A Framework of Deep Learning Classifiers to Diagnose COVID-19 in X-Ray Images (2020, methodology): “First, the cross validation estimator [43] was used and resulted in a confusion matrix as illustrated in Table 1.”
JSIS3D: Joint Semantic-Instance Segmentation of 3D Point Clouds With Multi-Task Pointwise Networks and Multi-Value Conditional Random Fields (2019, methodology): “Metrics include micro-mean accuracy (mAcc)1 [38] for semantic segmentation and mAP@0.”
Comprehensible classification models: a position paper (2014, background): “Comprehensible Classification Models a position paper Alex A. Freitas School of Computing University of Kent Canterbury, CT2 7NF, UK A.A.Freitas@kent.ac.uk ABSTRACT The vast majority of the literature evaluates the performance of classification models using only the criterion of predictive…”
Using Shapley additive explanations to interpret extreme gradient boosting predictions of grassland degradation in Xilingol, China (2021, background): “Recall is the proportion of correctly predicted positive examples in all observed positive examples (the observed NGD) (Sokolova and Lapalme, 2009). general, high-precision predictions have a low recall, and vice versa, depending on the predicted goals.”
Risky Action Recognition in Lane Change Video Clips using Deep Spatiotemporal Networks with Segmentation Mask Transfer (2019, background): “The evaluation focus of AUC is the ability for avoiding false classification [41].”
A Review On Evaluation Metrics For Data Classification Evaluations (2015, background): “Furthermore, data classification can be divided into binary, multiclass and multi-labelled classification [33].”
Optimal Thresholding of Classifiers to Maximize F1 Measure (2014, methodology): “Some commonly used measures for binary classification are accuracy, precision, recall, F1 measure, and Jaccard index [15].”
Deep Learning Enabled Fault Diagnosis Using Time-Frequency Image Analysis of Rolling Element Bearings (2017, methodology): “From this, the following 4 metrics were derived: precision, sensitivity, specificity, and 𝐹 -measure (see [37] for details on these metrics).”