Edit Clustering Query

To edit a Clustering Query Node, either double-click the node or right-click it and select Edit from the context menu. In the Select Columns window, specify the following information on the Cluster Predictions tab:

  1. On the Cluster Predictions tab, select the Case ID (optional)

    Case ID is optional, but it is recommended that you specify a Case ID to uniquely define each record. The Case ID helps with model repeatability and is consistent with good data mining practices.

  2. On the Cluster Predictions tab, specify the Number of Clusters to Compute. The default is 10.

  3. On the Cluster Predictions tab, edit Cluster prediction Output, as described in Edit Cluster Prediction Outputs. Data Miner automatically determines the output for the query. You can modify the output.

  4. On the Partition tab, add one or more Partition attributes (optional), as described in Add Partitioning Columns. Selecting a Partition attribute directs the predictive query to build a virtual model for each unique partition.

    To select partitions, click the Partition tab, and then click add targets. Use Add Partitioning Columns to select the partitions. You can also specify partitioning expressions.

  5. On the Input tab, modify Input (optional). You can add or remove inputs and change the mining types of inputs. Click Input. Make changes as described in Modify Input.

  6. On the Additional Output tab, add outputs (optional). By default, all target columns, the Case ID column, and partitioning columns are automatically added to Additional Output.To make changes, click Additional Output and follow the directions in Add Additional Output.