Creating Analytical Views

You can create a new analytical view if your dataset is defined through an Arcadia connection. Although analytical views are defined in the context of a dataset and its attributes, the metadata is saved to the data repository, alongside the base data tables. Therefore, an analytical view may persist even after its dataset is removed.

In this exercise, we are using the dataset National Geographic Features, based on the information found (mostly) on the United States Board on Geographic Names Website. We processed the archive AllStates_20160801.zip and added it as a single table, partitioned on the column state_alpha. We then created the dataset, and joined it with tables imported using feature-classes.csv and state-abbreviations.csv files.

The following steps show how to create 2 analytical views: geo.geo_features_lv_av_1 over a logical view, and geo.national_geo_features_av_1 over a base table.

  1. On the main navigation bar, click Data.

    The Data view appears, open on the Datasets tab.

  2. Under Datasets, find and click the dataset where you plan to define analytical views. We used National Geographic Features dataset.

    Remember that analytical views are only available on Arcadia connections.

    Selecting the Dataset for Analytical Views
    Selecting the Dataset for Analytical Views
  3. The Dataset Detail page appears.

    We strongly recommend that you create a logical view on this dataset, as described in Working with Logical Views.

    In this exercise, we created an Associated Logical View geo.geo_features_lv.

    Create an Associated Logical View
    Dataset Detail, with Logical View
  4. In the left navigation, select Analytical Views.

    Navigating to the Analytical Views Interface for the Selected Dataset
    Navigating to the Analytical Views Interface for the Selected Dataset
  5. On the Analytical Views page, click Create Analytical View.

    Creating a New Analytical View
    Creating a New Analytical View
  6. The Create Analytical View window modal opens. Note that by default, it shows the logical view as a foundation for building the new analytical view.

    In the Create Analytical View window modal, complete the following:

    • Under Select a base table/logical view for creating the Analytical View, leave the default: logical view geo.geo_features_lv.

      Note the basic format notation: databaseName.[logicalViewName|tableName].

    • Under Name of new Analytical View, enter the name of the analytical view. We used geo.geo_features_lv_av_1.

      Note that the naming convention for analytical views follows this pattern: databaseName.[logicalViewName|tableName]_av_n, where n is 1, 2, and so on.

    • Under Dimensions and Measures, select the items that are relevant to the queries supported by the new analytical view. Note that Partitions do not appear in the logical view, even though the primary table of this dataset uses partitions.

      We used Feature Class, State Alpha, and Record Count in this analytical view.

    • Click Create.
    Create Analytical View Modal Window; creating AV over LV
    Creating an Analytical View Over a Logical View
  7. On the Analytical Views page, click Create Analytical View, again.
  8. When the Create Analytical View window modal opens, select the primary table as a foundation for building this analytical view.

    In the Create Analytical View window modal, complete the following:

    • Under Select a base table/logical view for creating the Analytical View, select this table: geo.national_geo_features.

      Note the basic format notation: databaseName.[logicalViewName|tableName].

    • Under Name of new Analytical View, enter the name of the analytical view. We used geo.national_geo_features_av_1.

      Note that the naming convention for analytical views follows this pattern: databaseName.[logicalViewName|tableName]_av_n, where n is 1, 2, and so on.

    • Under Dimensions, Measures, and Partitions, select the items that are relevant to the queries supported by the new analytical view. Note that this time, Partitions appear in the logical view; because this view is based on a partitioned table.

      We used Feature Class, County Name, Record Count, and State Alpha in this analytical view.

    • Click Create.
    Create Analytical View Modal Window; creating AV over LV
    Creating an Analytical View Over a Partitioned Table
  9. The Analytical Views list for the dataset National Geographic Features appears.

    Notice that it shows the analytical views created in the previous steps, and lists them each below the name of the logical view or base table on which they are built.

    List of Analytical Views
    List of New Analytical Views

Notice also that you cannot use the analytical views geo.geo_features_lv_av_1 and geo.national_geo_features_av_1 at this time; they are defined, but not yet 'refreshed'. Therefore, the State field shows needs initial refresh.

You can see and manage all analytical views, across all datasets, in the general Analytical Views interface described in Managing Analytical Views.