Exploring a Data Source with Smart Insights

You can use smart insights for an at-a-glance assessment of your data source, and to quickly understand the information that its data contains.

The Prepare canvas provides two views of the data in your data source: Data view and Visual view. The Data view shows you a row-by-row snapshot of the data in the data source, however, it doesn’t help you determine how to best report on the data. The Visual view provides a visualization for each column, so you can quickly understand the distribution of the data in each column, including a row count for each attribute. The data with the most useful information is displayed at the top of the Visual view. To gain further information about your data, you can use the Summarize by field to show a specific measure's effect on the individual columns.

Note how Oracle Data Visualization presents information about the data source’s columns:

  • The most useful column information is presented first.

  • The type of visualizations shown is based on the column type. For text attributes, a horizontal bar chart is used. For date and time columns, a line chart is used. For numeric columns, a vertical bar chart is used.

  • Within a visualization, the most meaningful and useful values are shown.

  • You can mouse over a visualization to get more information about a specific aspect of a column’s data. For example, for the Product Category column, you can see the amount of revenue for each category, or for each region, you can see the number of rows or data.

  • You can analyze columns differently by using the Summarize by field to apply a measure to them. For example, if you summarize the data source by the Revenue measure, then you’ll see revenue by product name, revenue by state, revenue by city, and so on.

  • The number of bars shown in a horizontal bar chart depends on how the data is distributed. Typically ten bars are shown and all other data is displayed in a bar called Other. However, if 20% or more of the data falls into the Other bar, then the system breaks that data into the number of bars needed to give you a clearer picture of how the data is distributed. For example, if you’re working with a retail data source and you’re viewing the insights visualization for Sales by Order Month, and more than 40% of the sales happened in November and December, then the system adds two more bars to the visualization.

  • Based on the data, bins that represent ranges are shown. For example, if the column is Product Category, the visualization shows each product category based on number of rows using the 0, 100K, 200K, and so on bins.

Example of summarizing columns by a measure: You can use the Summarize by field to show the column values based on a specific measure. Note that in the following example the Summarize by field is set to Row count, which is the default:

Compare the preceding screenshot with the following one, which shows the Summarize by field set to the Profit measure. Note how the Visual view provides a different view of information contained in the columns.
To use smart insights:
  1. Create a new project or open an existing project.
  2. In the Project Editor, go to the Prepare canvas and click the Visual icon.
  3. In the Visual view, you can do the following:
    • Use the Summarize by field to select the measure that you want to apply to your columns.
    • Click the Options icon to show or hide null values in the visualization, or to include or hide the OTHER bar in horizontal bar chart visualizations.