Analytics
Decision Tree
Grows a CART decision tree that predicts a class (Gini) or a number (variance) from other columns, and writes the prediction and the readable rule behind it.
When to use it
Use it when you need to explain a prediction — interpretable segments such as “Plan = Basic and Tenure <= 17 → churns”.
In a pipeline
File InputDecision TreeSummarize (Multi-Aggregate)
Settings
| Setting | Type | Required | Default or hint |
|---|---|---|---|
| Predict (column) | select | Yes | Select column… |
| From columns (features) | select | Yes | Pick the columns that explain it |
| Kind of prediction | select | — | auto |
| Max depth | number | — | 4 |
| Smallest group (rows) | number | — | 5 |
| Prediction column name | text | — | Default: Prediction |
| Add the rule behind each prediction | checkbox | — | true |
| Add a Model column (JSON for Score Model) | checkbox | — | false |
Example
Who churns
- Predict (column)=Churned
- From columns (features)=Plan, Tenure
- Max depth=2
- Smallest group (rows)=1
| Plan | Tenure | Churned |
|---|---|---|
| Basic | 2 | yes |
| Basic | 30 | no |
| Pro | 3 | no |
| Basic | 4 | yes |
| Pro | 40 | no |
| Plan | Tenure | Churned | Prediction | Rule |
|---|---|---|---|---|
| Basic | 2 | yes | yes | Plan = Basic and Tenure <= 17 |
| Basic | 30 | no | no | Plan = Basic and Tenure > 17 |
| Pro | 3 | no | no | Plan ≠ Basic |
| Basic | 4 | yes | yes | Plan = Basic and Tenure <= 17 |
| Pro | 40 | no | no | Plan ≠ Basic |
Worth knowing
- Decide from the data predicts a number when the column holds more than 10 different numbers, and a class otherwise.
- Numeric columns split at a threshold; text columns split one value against the rest. Rows missing a value follow the larger branch.
- Max depth (default 4) and Smallest group (default 5) keep the tree readable and stop it memorising the data.
- Tick Add a Model column to hand the tree to Score Model; trees are not stored anywhere else.