The most well-known classification model evaluation metrics are:
- Confusion matrix,
- False positive rate,
- False negative rate,
- True negative rate,
- Negative predicted value,
- False dicovery rate,
- True positive rate,
- Positive predictive value,
- Accuracy,
- F beta score,
- ROC curve,
- ROC AUC score,
and so on.
Who studies the metrics? I am not quite clear about this question. Do you mean which is best amongst the evaluation techniques mentioned?
Stanley Ebhohimhen Abhadiomhen
The question can also be asked as follows: what is the best metric of classification models? Additionally: what is the difference between individual metrics?
Lamyaa Gamal El-deen Taha
Can you tell me a little more about these metrics and their properties?
Lamyaa Gamel El-deen Taha
Thanks a lot for the reply.
Here is the article:
https://neptune.ai/blog/evaluation-metrics-binary-classification
which speaks about many (24) metrics of classification. That confuses me. What are the differences? When are individual metrics applied?
- Badges
- Science topic
- Similar topics
- Mathematics
- Algebra
- Matrix