Immerse lets you narrow the data in your visualizations by clicking areas of the chart on which you want to filter.
You can reduce the scope of information to gain specific insights in your chart by filtering out values not pertinent to the analysis you are performing. You can create and name reusable filter sets in the Filter Panel. A filter set consists of filters that you define.
Every dashboard has a Default Filter Set. You can add filters to this default filter set, or you can create a new filter set by clicking the Filter set drop-down list box and then clicking Add filter set.
By default, any new filter set you create is named New Filter Set, followed by the number of the set created. You can edit the name of the selected filter set, duplicate it, or delete it by clicking the associated icon next to the filter set name in the drop-down list.
The Filters panel allows you to create logical filters based on values in your current data set. When you create or open a dashboard, the Filters panel is hidden by default. To show the Filter panel, click the arrow icon below the Filter set drop-down list box.
To hide the panel again, click the arrow icon on the right of the filter panel.
The number in parentheses to the right of the filter set indicates how many filters in the set are active.
If you have one or more defined filters, you can toggle individual filters on and off, or toggle the filter set on and off.
The following example creates a filter on a point map that restricts the selected records in the UFO Sightings data set to the contiguous United States.
Click the drop-down list box, and then click + Add filter set. The selected filter set should be New Filter Set 2.
Click the arrow icon below the Filter set drop-down list box to show the Filters panel.
Click + Add filter.
In the Data Selector, choose country.
In the criterion field, choose Exact match and enter US as the value.
Click + Add filter.
In the Data Selector, choose state.
In the criterion field, choose Does not equal and enter AK as the value.
Click + Add global filter.
In the Data Selector, choose state.
In the criterion field, choose Does not equal and enter HI as the value.
Click the drop-down list, and change the name of the filter set to Contiguous US.
The map displays UFO sightings only in the contiguous US. You can click the Default Filter Set to see all UFO sightings.
When you switch between filter sets, the URL in your browser window updates to include the current filter ID. If you copy and share the URL, other users can go directly to the dashboard, which is displayed with the same filter set.
If you use a URL that does not include the filter set ID, the dashboard displays with the last saved filter set. If the filter set is deleted, the dashboard displays with the default filter set.
To reduce the visual complexity of a Filter set, you can choose to display some or all of the filters in the set in Simple Mode. Every filter has a Simple Mode toggle (star) in the upper right of the filter box. Clicking this star toggle selects the filter for display when the overall Filters toggle is selected. For example, here the followees and followers filters are selected to be shown in Simple Mode, which also causes the View Simple Filters toggle in the upper right corner of the filter panel to be activated. The join_time and followers filters are not selected.
Clicking the View Simple Filters toggle hides the join_time and followers filters, and displays the followees and followers filters in Simple Mode:
Filters hidden in Simple Mode viewing are not removed or deactivated from the filter set. You can toggle filters on and off and adjust filters in Simple Mode.
If you do not want to apply a particular filter to all of the charts on your dashboard, you can apply that filter to a specific chart and, in some cases, change that filter.
To apply a chart-specific filter, open the chart for which you want to apply the filter; here, the chart # Records By Lang is selected. You can see that the current filter for the chart is country exactly matches US. You can change the filter, or click + Add filter to add another filter to the chart.
Because this filter uses Exact match, you can also configure quick filters for the chart that will be available in Dashboard view. Note that in the # Records By Lang chart above, the tab below the chart title reads US. Clicking the tab shows a drop-down searchable list of all countries available for the filter. Hovering over a country name (in this case, IN) in the list reveals an eye icon to the left.
Clicking the icon moves that filter to the top section of the drop-down list to indicate that it is selected, and makes that filter available to the chart when viewing the Dashboard. Here, quick filters have been enabled for FR, CA, IT, JP, US, and IN. IT is currently selected, and the chart information reflects that filter application. This makes it easy to apply high-priority filters to individual charts and see the results quickly, and do side-by-side comparisons of identical charts with different filters applied.
For line and histogram charts, filter by dragging left or right along the chart to select a range (brushing). The chart results are filtered to reflect only the highlighted range.
For Pointmap and Scatter Plot charts, filter by zooming in on regions. Hold the shift key to select a rectangular area of the map and zoom in. You can also use the mouse wheel or trackpad to zoom in and out.
Using Zoom Tools
You can focus on arbitrary sets of points in your Pointmap or Scatter Plot using the selection tools. See Zoom and Select.
Pointmaps also provide a Zoom To field, where you can zoom to a particular geographic location by name. You can enter a country, a city, or something as specific as a street address.
For other charts, click the section of the chart on which you want to focus.
Creating a Custom SQL Filter
You can create a custom SQL filter for fine-grained analysis of your data. The SQL filter is essentially the WHERE clause of a SQL statement.
For example, you might want to filter for UFOs that are either diamond-shaped or chevron-shaped.
Click the Filter set drop down list box, and then click + Add filter set.
Click + Add SQL filter.
In the editor, type
shape='diamond' or shape='chevron'.
Enter PointyShapes in the Name field.
Click Apply Filter.
The data set is now filtered to only display sightings of UFOs that are either diamond-shaped or chevron-shaped.
You can also choose to filter out certain data. Click the chart element you you want to suppress while holding down the Command key (Macintosh) or Control key (Windows/Linux). This feature is available for all chart types other than line, histogram, pointmap, and number.
At the dashboard level, when you filter results in one chart, the filter is reflected in any other chart on the dashboard that uses the same data table. For example, if you filter the state of Florida in a point map, all other charts on the dashboard adjust their results to only reflect records from the state of Florida.
To remove crossfilters, click the Clear crossfilters icon next to the selected Filter set.
Filters reduce the number of records displayed. Cohorts are named result sets of filtered data. Once you have named the cohort, you can work with those rows of data independently of the filters that created them, performing queries that span all rows in the data set.
For example, here is a dashboard depicting a very small (10-row) data set of 4 people who went to see 1 or more of 4 movies.
In the Cohort Builder, you can filter the results to focus on only the people who went to see the movie Speed. You can name the filtered result set Speed Lovers.
Since you filtered on the movie field, you normally would not be able to make a broader query to find out what other movies the 2 people who went to see Speed also went to see. But the cohort is now independent of its selection criteria. If you create a new filter using the cohort as a criterion, you can display records for all of the other movies seen by the fans of Speed.
Cohorts can include both global filters and aggregate filters. For example, you might create a cohort of beverages with an average price higher than $8.00.
Name and save the cohort and apply it to a new filter set.
Then you can use that cohort to look for a correlation between higher-priced beverages and higher-priced movie tickets.