Implementing drill-down in a dashboard

From IMSMA Wiki
Jump to: navigation, search

What does drill-down refer to?

In the context of data visualisation, "drill-down" refers to moving from one visualisation to another, usually from higher-level information to more detailed data. This is achieved by hyperlinking two visualisations and may involve passing parameters between the two visualisation data sets. Concretely, in MINT this would mean for example that a click on a specific part on a chart in a dashboard leads to another report or dashboard that details the data. As a simple example, one could have a chart representing the number of mine victims by gender. When clicking on either the "male" or the "female" part of the chart, the user would get to another chart, showing for example the number of victims by activity at the time of accident for the chosen category, i.e. male or female.

Principles

Drill-down is implemented through the following steps:

  1. Create an Ad-Hoc View with a chart containing a category that should be used for drill-down.
  2. Include that Ad-Hoc View on a dashboard
  3. Create a second Ad-Hoc View that will be the result of the drill-down. This Ad-Hoc View has to have a filter corresponding to the category defined in the first Ad-Hoc View (cf. step 1).
  4. Create a Report based on this second Ad-Hoc View
  5. On the dashboard created in step 2, right-click on the Ad-Hoc-View and select Properties, then go to the Hyperlinks tab. On this tab:
    1. Tick Enable chart links
    2. Under Action, select Open new page in order to open a new page/report/dashboard when drilling down
    3. In Repository URI, insert: repo:/<path_to_your_report>/<name_of_your_report>?<parameter_name>=$P{parameter}
    4. The parameter mapping will be created automatically.
Note.jpg It is a bit tricky to find out the exact parameter name. Refer to the example below for a trick on how to determine the exact name.

Example

Let's walk through a concrete example to illustrate the principle described above. For this, it is assumed to have a Domain on Victim data from IMSMA.

  1. Create a new dashboard
  2. On this dashboard, create a new chart dashlet (new content):
    1. Base it on a Victim data domain
    2. Create a very simple bar or pie chart to visualise the number of victims by gender
    3. Save the dashlet and the dashboard
  3. Create another simple Ad-Hoc View on the same domain to visualise the number of victims by activity at the time of accident (or by any other activity). Create a filter on Gender. Save as ad-hoc view and report and remember the name and location of the report.
  4. Go back to the dashboard created in the previous step
  5. Right-click on the victims by gender chart and select Properties
  6. Go to the Hyperlinks tab
  7. Tick Enable chart links
  8. Under Action, select Open new page
  9. In Repository URI, insert: repo:/<path_to_your _report>/<name_of_your_report>?genderenum_enum_1=$P{Gender}
    1. repo: means that the linked report is a resource from the MINT repository
    2. To determine the name of the parameter (in this case genderenum_enum_1), go to the drill-down Ad-Hoc View and hover over the filter attribute, in this case Gender and note the name appearing in the popup, such as genderenum_enum in the example below. Then, append _1 to this name to get the full parameter name, in this case genderenum_enum_1.
      MINT drilldown parameter popup.png
  10. Save the dashboard.
  11. Run the dashboard and click on different values (Male/Female) of your chart to experience the drill-down.