Add note for ordering multiclass business value matrix (#413)

NannyML · Jul 29, 2024 · 87f0d48 · 87f0d48
1 parent a954428
commit 87f0d48
Show file tree

Hide file tree

Showing 3 changed files with 15 additions and 2 deletions.
diff --git a/docs/how_it_works/business_value.rst b/docs/how_it_works/business_value.rst
@@ -55,6 +55,13 @@ observations in that cell of the :term:`confusion matrix<Confusion Matrix>`. Usi
 matrix notation the element on the i-th row and j-column of the business value matrix tells us the value
 of the i-th target when we have predicted the j-th value.
 
+.. note::
+ In Multiclass classification the classes are ordered alphanumerically.
+ This is used in the creation of the confusion matrix. The rows of the confusion matrix
+ represent target values in the corresponding alphanumerical order. And the columns
+ of the confusion matrix represent predicted classes in the same alphanumerical order.
+ Therefore the elements of the business value matrix should be constructed accordingly.
+
 For binary classification this formula is easier to manage hence we will use it as an example. Classificatio problems
 with more classes follow the same pattern.
 Using the `sklearn confusion matrix convention`_ we designate label 0 as negative and label 1 as positive.

diff --git a/...e_calculation/multiclass_performance_calculation/business_value_calculation.rst b/...e_calculation/multiclass_performance_calculation/business_value_calculation.rst
@@ -88,7 +88,10 @@ the following parameter specifications:
  The format of the business value matrix must be specified so that each element represents the business
  value of it's respective confusion matrix element. Hence the element on the i-th row and j-column of the
  business value matrix tells us the value of the i-th target when we have predicted the j-th value.
- It can be provided as a list of lists or a numpy array.
+ The target values that each column and row refer are sorted alphanumerically for both
+ the confusion matrix and the business value matrices.
+
+ The business value matrix can be provided as a list of lists or a numpy array.
  For more information about the business value matrix,
  check out the :ref:`Business Value "How it Works" page<business-value-deep-dive>`.
 

diff --git a/...ance_estimation/multiclass_performance_estimation/business_value_estimation.rst b/...ance_estimation/multiclass_performance_estimation/business_value_estimation.rst
@@ -80,7 +80,10 @@ parameters:
  The format of the business value matrix must be specified so that each element represents the business
  value of it's respective confusion matrix element. Hence the element on the i-th row and j-column of the
  business value matrix tells us the value of the i-th target when we have predicted the j-th value.
- It can be provided as a list of lists or a numpy array.
+ The target values that each column and row refer are sorted alphanumerically for both
+ the confusion matrix and the business value matrices.
+
+ The business value matrix can be provided as a list of lists or a numpy array.
  For more information about the business value matrix,
  check out the :ref:`Business Value "How it Works" page<business-value-deep-dive>`.