Evaluate Model Performance (Using the Lift Chart for Business Decisions…
Evaluate Model Performance
knowing functions more important than knowing technical details
Sample algorithm and model
Decision tree classifier
Many tree-based algorithms
Ranking high = not good
yellow and blue symbol = python
click decision tree classifier
scikit learn package
structured as an upside-down tree with the 9,494 patients
-Find the most predictive feature (the one that best explains the
target) and place it at the root of the tree.
-Split the feature into two groups at the point of the feature where
the two groups are as homogenous as possible
-Repeat step 2 for each new branch (box).
Click on the ROC Curve text in the XGBoost menu
Start with the orange oval on the
The ROC Curve, Figure 17.13,
sets the True Positive Rate against the False Positive Rate between the values of 0
and 1. The diagonal line shows the performance of a random assignment model. A
good model will tend to curve up toward the upper-left corner
Change the pulldown value to “Validation”
this change does not have a large
impact, and this is a good thing.
change the dropdown
value back to “Cross Validation.”
Using the Lift Chart for Business Decisions
Lift Chart pane on the XGBoost
The ideal scenario is having blue and orange lines that
are fully overlapping, indicating a strong model
Enable Drill Down
“Download” - ( ) to get a comma-separated values (.csv) file with all the cases and their predictions.