Before You Begin

This 15-minute tutorial shows you how to use location matches to check the accuracy of map results.

Background

You can view match results of the geographical locations in your map visualization. The Location Match dialog shows these types of matches:

  • 100% confidence
  • Multiple possible locations (ambiguous matches)
  • Low confidence with matches in the map layer (partial matches)
  • No location found in the map layer that matches your data point

You might want to:

  • Correct spelling of cities, states, or countries in your data source.
  • Add data such as county or province to resolve matches with multiple possible map locations.
  • Add filters to the map visualization to increase the match quality.
  • Create a custom map layer such as a postal code map layer that provides additional details about the data that isn't available in the default map layer.

    Creating a custom map layer requires administrator privileges and an understanding of map layers.

In this tutorial, you create a map visualization, correct and remove data, and create a filter using the available data.

What Do You Need?

Create a Dataset

In this section, you review location matches in your map visualization of sales by cities in the United States.

  1. Sign in to Oracle Analytics.
  2. On the Home page, click Create, and then select Dataset.
  3. In Create Dataset, click Drop data file here or click to browse, select the superstore_data.xlsx file, and then click Open.
  4. In Create Dataset From superstore_data.xlsx, click OK.
  5. Click Save Save icon. In Save Dataset As, enter superstore_data, and then click OK.


    Description of superstore_dataset.png follows
    Description of the illustration superstore_dataset.png

Create a Map Visualization

In this section, you create a map visualization with sales from cities around the world.

  1. Click Create Workbook. Close the Auto Insights panel.
  2. From the Data panel, hold down the Ctrl key, select City and Sales, and then right-click Pick Visualization. Select Map Map view icon.


    Description of city_sales_map.png follows
    Description of the illustration city_sales_map.png
  3. Click the visualization Menu Menu icon and select Location Matches.


    Description of city_sales_matches.png follows
    Description of the illustration city_sales_matches.png

    Location Matches shows 219 ambiguous matches because the United States includes multiple cities with the same name.

  4. Click OK.
  5. From the Data panel, drag State to Category (Location) in the Grammar panel to add more detail to the map.


    The map appears the same as the map visualization that just used City.

    Description of city_state_map.png follows
    Description of the illustration city_state_map.png
  6. Right-click the map visualization and select Location Matches.


    The addition of State to the map layer resolved all but 5 of the location match issues.

    Description of location_matches_state.png follows
    Description of the illustration location_matches_state.png
  7. Click Cancel.

Create a Filter

In this section, you use a filter in the visualization to narrow your map view to a specific area of interest, and possibly reduce the number of location mismatches.

  1. In the canvas, click Add Filter Add Filter icon. From superstore_data, select Region. From Region, select East, and then click outside of the selection dialog.


    The map visualization changes to show the data points in the East region. However, a couple of data points appear in the mid-west.

    Description of region_filter_map.png follows
    Description of the illustration region_filter_map.png
  2. Click the visualization Menu Menu icon and select Location Matches.


    The region filter reduced the location match issues to four. If you added county to your dataset, the out of region location match issues might disappear. Because the City of Orange, New Jersey, and East Orange, New Jersey are in the same county, the location match issues probably won't totally disappear without a custom map layer of New England cities. A custom New England cities layer would also resolve the Darien and Fairfield, Connecticut location match issues.

    Description of region_filter.png follows
    Description of the illustration region_filter.png
  3. Click Save. In Save Workbook, enter Cities Location Matches in Name, and then click Save. Click Go back Back icon to return to the Home page.

Create a Map Visualization with World Cities

In this section, you create a new workbook with a different data source.

  1. On the Home page, click Create, and then select Dataset.
  2. In Create Dataset, click Drop data file here or click to browse, select the world_cities_data.xlsx file, and then click Open.
  3. In Create Dataset table from world_cities_data, click OK.
  4. Click Save Save icon. In Save Dataset As, enter world_cities_data, and then click OK
  5. Click Create Workbook. Close the Auto Insights panel.
  6. From the Data panel, hold down the Ctrl key, select City and Sales, and then right-click Pick Visualization. Select Map Map view icon.


    Description of city_sales_world.png follows
    Description of the illustration city_sales_world.png
  7. Click the visualization Menu Menu icon and select Location Matches.


    Location Matches can't find the small French village of Eze and Riccione, Italy on the default map.

    Description of world_cities_loc_matches.png follows
    Description of the illustration world_cities_loc_matches.png
  8. From the Data panel, drag Country to Category (Location) in the Grammar panel to possibly improve matches.


    Description of city_country_map.png follows
    Description of the illustration city_country_map.png
  9. Click the visualization Menu Menu icon and select Location Matches.


    With the addition of Country to the world cities map layer, the village of Eze has four matches and Riccione has 22 matches. The number of location match issues are reduced from 21 to 16.


    Description of country_city_world.png follows
    Description of the illustration country_city_world.png
  10. In Location Matches, click the check box in the Riccione row, and then click Remove to create a filter to remove the selected item. Click OK.
  11. Right-click the map view, and then click Location Matches.


    Adding country removed some of the ambiguous matches, and Riccione no longer appears in Locations Matches. In your workbook, scroll through the dialog to review the location matches.

    Description of riccione_removed.png follows
    Description of the illustration riccione_removed.png

Correct Data Source Issues

In the Location Matches, under the Your Data column, the City of Caiyro is spelled incorrectly.

  1. In the workbook, click Save. In Save Workbook, enter World Sales in Name, and then click Save.
  2. Open the world_cities_data.xlsx file. In the spreadsheet, use Ctrl + F, and then in Find What, enter Caiyro. Close the Find dialog.
  3. In the row with Caiyro, enter Cairo, click Save, and then close the file.

Update the Data in Your Workbook

In this section, you reload the updated data source and refresh the workbook.

  1. On the Home page, click Data in the search bar, enter world_cities_data, and then click Search. In world_cities_data, click Actions Menu icon, and then select Reload Data.
  2. On the Home page, click Workbooks, enter World Sales, and then click Search.
  3. In the World Sales workbook, click the Actions menu icon, and then select Open.
  4. In the workbook, click Menu Menu icon, and then select Refresh Data.
  5. Click the visualization Menu Menu icon and select Location Matches.
  6. In Location Matches, scroll to Cairo to view the correction made in the dataset.


    Description of cairo.png follows
    Description of the illustration cairo.png

Create a Pivot Table

In the Location Matches dialog, some data points had multiple matches. In this section, you create a pivot table that enables viewing all of your City sales data points.

  1. In the Data panel, hold down the Ctrl key, and select the following:
    • Country
    • City
    • Sales
  2. Right-click, select Pick Visualization, and then select Pivot Pivot view icon.
  3. In the Grammar panel, drag City to Color, and then drag County to Rows.

    You can scroll through the list in the pivot table to review the data.

    Description of city_country_sales_pivot.png follows
    Description of the illustration city_country_sales_pivot.png

Learn More