Altair SmartWorks Analytics

 

Aggregating Data

The Aggregate node allows you to aggregate and group data in tables.

If you had a report containing the details shown below, for example:

 

you may wish to group the data by Customer and Media Type so that you obtain a report that clearly describes the total sales each customer generated for a specific media type.

  

Prerequisite

  • An Execution Profile with an active session linked to the workflow

Steps

  1. Produce a Data Frame node by importing a CSV or database table.

  2. From the Data Preparation group of  the Nodes tabbed page, drag and drop the Aggregate node from the Palette to the Workflow Canvas. The Aggregate node has one input socket and one output socket. Connect the output socket of the Data Frame node to the input socket of the Aggregate node.

  3. Configure the Aggregate node by opening the Node Viewer. You can double-click on the node or use the Open option provided in the node menu. 

  4. The Configuration tab displays by default. Specify the following details to configure the Aggregate node.

  5.  

    Property

    Description

    Input properties

    The input table name for the Aggregate node. This field is non-editable.

    Output properties

    The name of the output table of the Aggregate node.

    Aggregation Settings: Case Sensitive

    Specifies whether records should be grouped together if their values in grouping columns have identical case. When the check box is selected, "Mary" and "Mary" will be grouped together but "Mary" and "mary" will not.

    Group Columns

    Specifies which columns contains values that should be grouped. Click on the Group icon beside a grouping column in the Columns to Include section. This column is moved to the Group Columns section of the Aggregate Configuration panel.

     

    Columns added to the Group Columns section are automatically displayed in the output table.

    Columns to Include

    Specifies which columns contain values to be aggregated.

    Select the columns containing values to be grouped by checking its box, click f(x) next to the column name, select the appropriate aggregation function to apply (e.g., Sum, Minimum, Maximum, Average) and then click Apply. The default function applied to numeric columns is Sum, and the default function applied to text and date/time columns is Count.

     

    The data preview table/data grid is updated according to your aggregation specifications.

    Output columns

    Specifies which other columns should be included in the output table. By default, all of the columns are selected.

     

    Clear the check box next to each column to exclude this column from the output columns.

    Included column names

    Select the prefix or suffix for column names to be included in the aggregation. 

     

     

  6. Click Save to save the configuration settings.

  7. The auto generated code is displayed in the Code tabbed page. The Input Properties section is read-only and pre-populated with the input table name. However, the Output Properties section allows you to modify the configured properties.

     

  8. Click Save to save the changes that you have made and then click Run to execute the Aggregate node. 

  9. Alternatively, click Discard to discard the changes that you have made.