2017 International Conference on Computer Science and Engineering (UBMK) 2017
DOI: 10.1109/ubmk.2017.8093539
|View full text |Cite
|
Sign up to set email alerts
|

Binary classification performance measures/metrics: A comprehensive visualized roadmap to gain new insights

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
30
0
1

Year Published

2018
2018
2024
2024

Publication Types

Select...
4
3
1

Relationship

0
8

Authors

Journals

citations
Cited by 70 publications
(31 citation statements)
references
References 13 publications
0
30
0
1
Order By: Relevance
“…Four performance evaluation metrics are considered, including Accuracy, Sensitivity, Specificity and Area under the Receiver Operative Curve (AUC). These are all derived from a contingency table [28]. In this case, the Accuracy measures the total number of correct predictions.…”
Section: Resultsmentioning
confidence: 99%
“…Four performance evaluation metrics are considered, including Accuracy, Sensitivity, Specificity and Area under the Receiver Operative Curve (AUC). These are all derived from a contingency table [28]. In this case, the Accuracy measures the total number of correct predictions.…”
Section: Resultsmentioning
confidence: 99%
“…As defined by literature [39], the following four metrics are the optimum base measures used to assess the performance of binomial classification models. These base measures are calculated from a series of experiments using a set of positive and negative instances, where TP = true positive, FP = false positive, FN = false negative, TN = true negative.…”
Section: Methodsmentioning
confidence: 99%
“…Due to the four complementation relations within pairs of these eight measures, the number of independent quantities among them is at most four. It seems that there is a widespread (and at least implicit) belief that this number is exactly four (usually obtained by counting the four direct indicators , , and ) [12][13][14][15][16][17][18][19][20][21][22][23][24] . We show in Section 4 that this number is, in fact, three, by simply being able to express any of the four direct indicators in terms of the other three.…”
Section: On Diagnostic Testing and Its Basic Measuresmentioning
confidence: 99%
“…We use equations (13)(14)(15)(16) to express each of the four most prominent indicators of diagnostic testing (Specificity, Negative Predictive Value, Sensitivity, and Positive Predictive Value) solely in terms of the other three, namely…”
Section: Formula Derivation Via a Novel Signal Flow Graphmentioning
confidence: 99%