Each rules table displays the top rules and values for confidence, rule support, lift, condition
support, and deployability. Each table is sorted by values of the selected criterion. You can
display all rules or the top Number of rules, based on the selected
criterion.
Sortable Word Cloud
A list of the top rules, based on the values of the selected criterion. The size of the text
indicates the relative importance of the rule. The interactive output object contains the top rules
for confidence, rule support, lift, condition support, and deployability. The selected criterion
determines which list of rules is displayed by default. You can select a different criterion
interactively in the output. Max rules to display determines the number of
rules that are displayed in the output.
Map
Interactive bar chart and map of the top rules, based on the selected criterion. Each
interactive output object contains the top rules for confidence, rule support, lift, condition
support, and deployability. The selected criterion determines which list of rules is displayed by
default. You can select a different criterion interactively in the output. Max rules to
display determines the number of rules that are displayed in the output.
Model Information Tables
Field Transformations.
Describes the transformations that are applied to fields used in the analysis.
Record Summary.
Number and percentage of included and excluded records.
Rule Statistics.
Summary statistics for condition support, confidence, rule support, lift, and deployability. The
statistics include mean, minimum, maximum, and standard deviation.
Most Frequent Items.
Items that occur most frequently. An item is included in a condition or a prediction in a rule.
For example, age < 18 or gender=female.
Most Frequent Fields.
Fields that occur most frequently in the rules.
Excluded Inputs.
Fields that are excluded from the analysis and the reason each field was excluded.
Criterion for Rules Tables, Word Cloud, and Maps
Confidence.
The percentage of correct rule predictions.
Rule Support.
The percentage of cases for which the rule is true. For example, if the rule is "If x=1 and y=2,
then z=3," rule support is the actual percentage of cases in the data for which x=1, y=2, and z=3.
Lift.
Lift is a measure of how much the rule improves prediction compared to random chance. It is the
ratio of correct predictions to the overall occurrence of the predicted value. The value must be
greater than 1. For example, if the predicted value occurs 20% of the time and the confidence in the
prediction is 80%, then the lift value is 4.
Condition Support.
The percentage of cases for which the rule condition exists. For example, if the rule is "If x=1
and y=2, then z=3," condition support is the proportion of cases in the data for which x=1 and
y=2.
Deployability.
The percentage of incorrect predictions when the conditions are true. Deployability is equal to
(1-confidence) multiplied by condition support or condition support minus rule support.