This tool allows users to explore relationships between objects through network analysis. The tool provides a data stream that users can output and use to build a network visualisation in Tableau.
The ‘Plot Network’ tool requires two inputs. The first input ‘N’ or ‘Node’ should represent a unique list of each node (object) that exists within your network.
The node data stream may also include other columns which contain additional attributes and characteristics about a node, for example a group field. Such fields will be passed into the output.
The second stream, the ‘E’ or ‘Edges’ stream, represents the relationships that exist between the nodes given in the node data stream.
The data must contain at least 2 fields, one field representing the ‘from’ and a second field representing the ‘to’. Again additional attributes may be included and these will be passed into the output.
Data should be aggregated so each relationship only exists once though a relationship between two objects can be in two directions and thus have two lines.
The configuration pane consists of a single tab with four questions.
Select node field
The node field is the field from your ‘N’ or ‘Node’ data stream which represents the unique values or names of the different objects within your network.
Select from field
The from field is the field from your ‘E’ or ‘Edge’ data stream which represents the node from which the relationship starts.
Select to field
The to field is the field from your ‘E’ or ‘Edge’ data stream which represents the node at which the relationship ends.
Select layout algorithm
This field contains a drop down of the different layout algorithms that exist within network visualisations. This paper highlights the methodology behind some of the layout algorithms that exist within the drop down.
The second output, ‘N’ or ‘Node Summary’ output provides you with the input data to the ‘Node’ stream enriched with centrality metrics about how important each node is to the network.
There are 5 centrality metrics reported in this output, betweeness, closeness, degree, evcent and pagerank. This article gives detail on the centrality metrics included in the report.
The ‘D’ anchor, or ‘Data’ anchor output, provides users with a datastream which would allow them to build a network visualisation in Tableau. The data is structured using this method.
An example visualisation can be found in the ‘Template Visualization’ section which you can download and view how the data can be used to build a network visualisation in Tableau.
Users can find an example workflow containing the ‘Plot Network’ macro here.
An example visualization which uses the ‘D’ or ‘Data’ output stream from the macro can be found here. Users can download this visualisation to understand how the data can be used to build a network visualization in Tableau.
Users can also use this technique to replace the datasource used in the template with their output and Tableau will automatically refresh and produce the visualization with their data.