You are here:
Explore Predictions for Multiclass Outcomes
Explore predicted outcomes for multiclass models. Perform interactive, “what if” analyses and change feature selections to see the winning prediction, ranked list of outcomes by probability, model accuracy, and top prediction factors.
Multiclass classification models predict probable outcomes for 3–10 categories. For example, a manufacturer can predict, based on customer attributes, which of three service contracts a customer is most likely to choose. In a multiclass model, a prediction selects the class with the highest probability of winning based on the values of predictor variables. Knowing the most influential predictor variables can, for example, help a salesperson anticipate the most probable outcome in a given circumstance.
To simulate and explore future outcomes:
- Open a Model that is associated with a multiclass classification use case.
-
On the Model Toolbar, click Predictions. The screen provides
selectors in the left panel.
-
Under Select a Group to Predict, select an option from one or
more features. The right panel shows the predicted results.
# Field Description 1 Einstein Prediction Prediction score for your selections. Probabilistic calculation of the outcome based on the multiclass classification model. 2 Ranked List Chart that shows the predicted probabilities of each class, starting with the winning class (blue bar). 3 Model Accuracy Chart that shows the accuracy of the model as applied to the training dataset. 4 Top Prediction Factors Explanatory variables, favorable and unfavorable, that are most strongly associated with the predicted outcome. 5 Predicted Outcome Chart that shows how the top predictors influenced the predicted outcome.
