A Simple Generalisation of the Area Under the ROC Curve for Multiple Class Classification Problems
Citations
17,017 citations
Cites background from "A Simple Generalisation of the Area..."
...Hand and Till (2001) take a different approach in their derivation of a multi-class generalization of the AUC....
[...]
... Hand and Till (2001) take a different approach in their derivation of a multi-class generalization of the AUC....
[...]
...Hand and Till (2001) point out that Gini + 1 = 2 · AUC....
[...]
...Hand and Till (2001) point out that Gini + 1 = 2 · AUC....
[...]
6,320 citations
Cites background or methods from "A Simple Generalisation of the Area..."
...Interested readers can refer to [ 131 ] for a more detailed overview of this technique....
[...]
...To eliminate this constraint, Hand and Till [ 131 ] proposed the M measure, a generalization approach that aggregates all pairs of classes based on the inherent characteristics of the AUC....
[...]
...Similarly, under the multiclass imbalanced learning scenario, the AUC values for two-class problems become multiple pairwise discriminability values [ 131 ]....
[...]
3,672 citations
Cites methods from "A Simple Generalisation of the Area..."
...When there are multiple classes, the extensions of ROC curves described by Hanley and McNeil (1982), DeLong et al....
[...]
3,406 citations
2,046 citations
Cites background from "A Simple Generalisation of the Area..."
...Hand and Till (2001) take a different approach in their derivation of a multi-class generalization of the AUC....
[...]
...Hand and Till (2001) point out that Gini + 1 = 2 × AUC....
[...]
References
37,183 citations
21,694 citations
19,398 citations
"A Simple Generalisation of the Area..." refers background or methods in this paper
...Amongst the most popular are misclassification (or error) rate, and the criterion with which this paper is concerned, the area under the Receiver Operating Characteristic (ROC) curve (e.g., see Hanley & McNeil, 1982; Zweig & Campbell, 1993; Bradley, 1997)....
[...]
...This relationship is well-known (e.g., see Hanley & McNeil, 1982 ), but both it and the implications for estimation seem not always to be appreciated, so we describe it here for completeness....
[...]
...Referring to Hanley and McNeil (1982) we obtain the standard error of ˆ...
[...]
...Amongst the most popular are misclassification (or error) rate, and the criterion with which this paper is concerned, the area under the Receiver Operating Characteristic (ROC) curve (e.g., see Hanley & McNeil, 1982; Zweig & Campbell, 1993; Bradley, 1997)....
[...]
...Referring to Hanley and McNeil (1982) we obtain the standard error of  to be se( Â) = √ θ̂ (1 − θ̂ ) + (n0 − 1)(Q0 − θ̂2) + (n1 − 1)(Q1 − θ̂2) n0n1 (6) with θ̂ = S0 n0n1 and Q0 = 1 6 (2n0 + 2n1 + 1)(n0 + n1 + 1)(n0 + n1) − Q1 where Q1 = n0∑ j=1 (r j − 1)2....
[...]
14,825 citations
12,940 citations
"A Simple Generalisation of the Area..." refers methods in this paper
...All except the CHROMOSOME data set were obtained from the UCI Repository of Machine Learning Databases ( Blake & Merz, 1998 )....
[...]