Einstein Discovery - Explore Story Insights pt. 2 Flashcards
How to Compare Subgroups with the Global Average?
Descriptive
See how a category compares to variable’s global average.
Open story, variables panel, select variable you want to investigate.
Then, filter selector, choose value you want to compare with global average
Explore WHY a value in a variable does better or worse than average. Also uses waterfall chart. what is that?
Diagnostic
Waterfall chart -
Visualization of the factors, how they contributed to the predicted outcome, and in comparison to the global outcome.
X-axis represents the drivers’ contribution to the story’s outcome.
Y-axis represents the drivers (explanatory variables) of the outcome variable.
What elements are in a waterfall chart?
Global Average - gray bar
Driver (First Impact) -
Category selected for this insight. Is the second bar from the top.
Next Drivers
Small terms related to -
aggregated effect of all terms interacting with the category that do not appear in the other bars shown previously
Other Impacts -
Terms that are not specific to Category but that still occur more or less often.
Unrelated small Contributors -
not specific to Category. This section shows us factors that have positive or negative effects on all Outcomes.
Unexplained -
quantifies the gap that this model hasn’t been able to attribute to different drivers.
Category Average -
Blue bar, avg for selected category
Waterfall table, how do impact and contribution to outcome relate?
Mean the same thing.
The value represents an estimate of the contribution of a single term to the predictions in aggregate, across all predictions. It compares the subset to the global dataset.
What happens when you compare subgroups in ED
Einstein shows a summary and waterfall chart of the comparison between the selected values.
Subgroups waterfall First Explanatory Variable
The gray bar represents the first (left-most) explanatory variable selected in the story toolbar. Hover over this bar to see additional information.
Subgroups waterfall Second explanatory Variable
The blue bar represents the second explanatory variable selected in the story toolbar. Hover over this bar to see additional information.
Subgroups waterfall green bars represents what?
Green bars reflect cases in which explanatory variables have a favorable effect on the outcome. A favorable effect moves you closer to your story’s goal, such as increasing opportunity wins when the goal is to maximize opportunity wins.
Subgroups waterfall red bars represent what?
Red bars reflect cases in which explanatory variables have an unfavorable effect on the outcome. An unfavorable effect moves you away from your story’s goal, such as increasing customer churns when the goal is to minimize customer churns.
subgroups waterfall small terms show what?
Represents the aggregate of terms that are too small to include in the other categories. Therefore, these terms are combined into a single set of statistics. Hover over this bar to see additional information.
subgroups waterfall unexplained shows what?
Represents the unexplained variables on the Outcome. Hover over this bar to see the effect of unexplained variables on the outcome.
What does explore predictions and improvements entail?
Explore predicted outcomes and suggested ways to improve those predicted outcomes. Perform interactive, “what if” analyses and change feature selections to see prediction scores, top prediction factors, and top improvements to enhance prediction scores
What is an improvement?
An improvement is a suggested action, based on prescriptive analytics, that a user can take to improve the likelihood of a desired outcome. Improvements are associated with actionable variables, which are explanatory variables that people can control
How to get to Predictions in a story?
Open Story, select Predictions on story toolbar at the top
What fields show up when you select a group to predict in Predictions?
Einstein Prediction - Prediction score for your selections. Probabilistic calculation of the outcome based on the story’s model.
Top Improvements - Suggested actions that you can take to improve the predicted outcome.
Model Overview - Metrics that describe the quality of the model
Top Prediction Factors - Explanatory variables, favorable and unfavorable, that are most strongly associated with the predicted outcome.
Insights - Any insights associated with your prediction.
