Timeline for Determinant of confusion matrix: how useful is it as a performance metric?
Current License: CC BY-SA 3.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| May 10, 2023 at 23:36 | answer | added | BradLipovsky | timeline score: 2 | |
| Feb 20, 2022 at 0:23 | answer | added | Arkadi | timeline score: 2 | |
| Aug 20, 2017 at 4:20 | answer | added | andraugust | timeline score: 2 | |
| Jul 15, 2016 at 18:15 | comment | added | Firebug | @whuber The "basis" is it's actually being used. The example is analogous to accuracy (you can find questions about the flipping procedure here on stats). I expect answers about if it can be useful in multiclass classification, imbalanced classes, or any other application out of the idealized binary, equal sized, classes. | |
| Jul 15, 2016 at 18:03 | comment | added | whuber♦ | This is a strange question. You suggest a procedure without supplying any basis for it, offer an example showing how awful it can be, and then ask "is it useful?" What kind of answers do you expect? | |
| Jul 15, 2016 at 17:06 | history | asked | Firebug | CC BY-SA 3.0 |