Please enable JavaScript.
Coggle requires JavaScript to display documents.
Chapter 20 Larsen (There are four kinds of features to consider before…
Chapter 20 Larsen
-
It is important to distinguish between information useful for understanding the model and information useful to an audience for making business decisions.
Understanding the model is important for the data scientist creating and modifying the model, but a seldom few model details are relevant to the final stakeholder audience.
-
2.Model quality metrics (confusion matrix)
3.Areas where a model struggles (potential for improvement through more data––features & cases)
4.Most predictive features for model building
5.Feature types especially interesting to management (e.g., insights into the business problem and unknowns uncovered during the modeling process)
6.Recommended business actions (i.e., to implement model or not, any business decisions to implement at various probability thresholds, and how will doing so change practice?)
All algorithms being worked with function conceptually in the same way: they determine the generalizable relationship between the features (or independent variables, if your audience is used to statistics)
and the target (or dependent variable)and place those relationships into a model that can be used to both understand those relationships and predict the outcome of cases not yet encountered.
Make sure there haven't been any mistakes made in the model creation process. This is done by releasing the holdout data.
it was found that a small set of data can sometimes lead in the wrong direction. The holdout sample provides an opportunity to check whether such problems may have occurred.
but it is best to restrict this data processing information to one slide without code when presenting the findings
begin with the confusion matrix for the chosen model and annotate it for the audience, per
it's possible that performance will deteriorate if the model is given new data but that this can be monitored, and, in the case of reduced performance, the model can be retrained when needed
-
-
Often information that helps a model make high-quality decisions is not helpful in changing practice.