Open Settings

How to use CubeViz

CubeViz is a facetted browser for statistical data utilizing the RDF Data Cube vocabulary which is the current state-of-the-art regarding the representation of statistical data in RDF. This vocabulary, which is compatible with SDMX and increasingly being adopted. Based on the Data Cube vocabulary, CubeViz generates a facetted browsing widget that provides possibilities to interactivly filter observations which are to be visualized as charts.

The following sections briefly sketches how to use CubeViz:

How to select Data Source and Dataset

Cubeviz is build upon the Semantic Web Technology Stack and is able to use Linked Open Data Stores for querying Statistical Data Sets. On the upper left corner the user can open the settings Dialogue (Open Settings Button) offering the ability to switch between pre-configured stores. If a store is chosen from the list, the possibly currently open dataset become deselected and the user get presented a list of available datasets in this store. After selecting one of the datasets from the list, CubeViz analyzes the dataset against some integrity constraints to check if it contains statistics and if they are well represented.

If the dataset contain well represented statistics, CubeViz will answer with a respective message. Below the success message a button is visible that will start CubeViz. If the dataset is not processable by CubeViz, a message will appear. In this case, you can select an other dataset or contact the data or tool provider.

How to search in a Dataset and how to share a Selection

As a starting point in order to browse statistical data using CubeViz, users should focus on the Data Selection Box located on the upper left part of the window. This box is shown in Fig.2 as well. Following the structure given by the DataCube vocabulary, users are able to select specific structural elements of the DataCube.

After selection of the desired Statistical Data Set, CubeViz is requesting the set of dimensions with respect to the selection. Below the dimension labels, the count of selected dimension elements is listed (first number) together with amount of all available dimension elements. A Change link located on the right hand of each listed dimension opens a dialogue when clicking.

This dialogue (see Fig.3) offers possibilites to manually select dimension elements that the user are interested in. After selecting the desired dimension elements, users should click the Update Selection button on the bottom right corner of the dialogue. Behind the dimension labels the count of selected dimension elements should be updated and the Chart will be updated immediately.

In order to decrease the amount of server requests, CubeViz processes the output mostly on the client side (in the browser) and often re-uses already received data parts. As a result of this behavior, the address bar of the browser is not being updated if users make selections. However, users are provided with a link of their selection, if they click the Share button (bottom left corner of the Data Selection Box). Subsequently they can either directly use the generated link or copy it to the clipboard using the context menu of the browser.

Selection of suitable Charts

According to the user selection of structural elements (see section above) and the resulting set of observations, CubeViz offers suitable chart types. The first suitable chart is selected automatically and is displayed directly after clicking the button Show/Update Visualization. Further suitable charts can be selected on the top right corner of the application.

Not all chart types are suitable for all individual user selections. A pie chart,for instance, can be used to compare different observation on the basis of elements of one dimension (see Fig.4). Furthermore spline, column and a bar chart are used to compare observations on the basis of elements of more than one dimension (See Fig.5 , 6 and 7). On the page of examples, a set of samples are listed, explaining the successive processes CubeViz executes for the respective selection. The last option in the chart switcher is a table, which is lists all resources, that are on the one hand selected by users (DataStructures, Datasets, Dimensions and Dimension elements) and on the other hand the observations which are linked to them. All of these resources are linked to their source location, which are usually outgoing links.

Selection of Chart options

Most of the charts offer options that help users to optimize the output according their needs. These options can be set up in the Chart Options Box as illustrated in the depictions below. To open the chart option box, users only have to click on the chart icon again. After opening this box, users can select exactly one element per identifier and proceed with the selections while clicking on the button Update chart. To close this box without accepting changes, the specific button on the top-right corner of the box can be clicked.

In the following we sketch available options:

  1. Type: Some of the charts offer the selection of a respective subtype. The line chart output, for instance, can be refined by setting up the subtype spline or scatter-plot. The polar chart output is created with the default subtype column and can be refined with the options column, spline, line, area, areaspline and scatter.
  2. Switching axes: If users create a selection of dimension elements along two dimensions it is sometimes necessary to switch the relation between dimensions and its regarding axis. Comparing Fig.7 and Fig.8, which display exactly the same observations, the difference is visible and the benefit of this option conceivable.
  3. Data Labels: This option can be set up if the user wants to see the underlying measurement values as part of the chart output. The measurements are shown by default as part of the tooltip, which is opens while hovering over observations.
  4. Stacking: If users select many dimension elements, the lucidity can be lost. This option can help to get the lucidity back as can be seen in Fig. 5 or 9.
  5. Scale: The scale of a chart is set up as linear by default. If dimension elements are to be compared (or its regarding observations) have a hughe divergence of its measurements, it is possible that small values can not be displayed. To display them the scale can be set up to logarithm.
  6. Inverted: This option is different to the option of Switching Axes. While setting up the option inverted the dimensions elements on the X-Axis are switched with them from the value axis.


Sparql Endpoints

Select a datasource (SPARQL endpoint):

or add a data source (SPARQL endpoint)


 Models of selected SPARQL endpoint