Machine Learning Basic Course
ms19
Dataiku DSS Core Designer, Dataiku DSS ML Practitioner, Dataiku DSS Adv Designer, Registered Posts: 1 ✭
Hi,
I have been following the ML Practitioner path. In the Machine Learning Basic course-> Tune the model section->Hands on segment, there is a following statement:
"A business analyst has analyzed the relationship between the top two variables from the Variable importance chart, age_first_order and pages_visited_avg, and the target, high_revenue, to assert the following:
- When age_first_order is greater than or equal to 40, the customer is likely to be labeled “high revenue = true” at least 10% of the time.
- When count of pages_visited_avg is between 6 and 12, the customer is likely to be labeled “high revenue = true” at least 10% of the time."
My question is, where can you check explicit values of variables? I can only see % in the Variable Importance section in the model details.
Tagged:
Answers
-
Hi,
In fact, you can’t assert that directly from the Variable importance chart. However, you can use:
- the subpopulation analysis, where you have the confusion matrix for different ranges of values of a chosen feature.
- the partial dependence panel to understand the relationships between a feature and the target variable.
Metrics & assertions also do this but you have to retrain the model.
I hope this will help.
Isabelle