Which of the following is most likely happening?

University of Michigan - Ann Arbor

Recall that the formula for precision is $Pr = TP / (TP + FP)$. To compute the macro-average precision of a multi-class classification, compute $Pr$ for each class, add them together, and then divide by the number of classes. We can summarize this as the following formula:

macro-average precision = $(\sum_{i=1}^nPr_i) / n$, where $Pr_i$ is the precision of the $i$th class and there are $n$ classes.

Macro-average Precision

We can compute micro-average precision by dividing the number of true positives across all classes by the sum of true positives and false positives across all classes.

More formally, let $TP_i$  and $FP_i$ be the number of true positives and false positives, respectively, in class $i$. Given a multi-class classification with $n$ classes, we can express micro-average precision as the following formula:

micro-average precision = $(\sum_{i=1}^nTP_i) / (\sum_{i=1}^n(TP_i+FP_i ))$

Micro-Average Precision

https://towardsdatascience.com/top-5-metrics-for-evaluating-classification-model-83ede24c7584

Classification Evaluation Metrics

A classifier is trained on an imbalanced multi class dataset. After looking at the model's precision scores, you find that the micro averaging is much smaller than the macro averaging score.

Learn Before

Related