Argument Explorer

../_images/OWArgExplorer.png

Network visualization of argument attacking relationships.

Inputs

  • `Edge Data: Data table that contains edge information of the argument attacking network, including columns: source, target, weight.

  • Node Data: Data table that contains node information of the argument attacking network.

Outputs

  • Selected Data: Data table that contains information of the selected nodes.

Description

Argument Explorer has the following function:

  • Network visualization: The argument attacking network is visualized with node coler representing their labels (green for supportive and red for defeated) and edge width for showing weights.

  • Node selection: This widget allows to select node(s) and this will update the output table that contains the information of selected nodes. Also, when a node is selected, all the edges relevant to that node will be highlighed by hiding the unrelevant edges.

  • Layouting: A set of network layout can be chosen, that include spring, multipartite, kamada kawai, and spectral.

  • Navigation: This widget supports a series of navigating functions for better observing the network, that include zooming, panning, and centralizing. Also, by hovering over a node, the relevant meta information of that node will be shown in the popping-up tooltips.

Control

  • Graph layout: Layout used for positing nodes and edges in the network.

  • Node sparsity: Spatial closeness of nodes, in range of [1, 10]

  • Zoom/Select: Navigation tools for better observing the network.

  • Send Automatically: if the checkbox is enabled, the information of selected nodes will be automatically sent to the output data.

Example

Here is an example workflow that shows how the widget works:

../_images/wf_explorer.png

where the input Edges and Nodes table look like this:

../_images/df_edges.png ../_images/df_nodes.png

The result network can be observed directly from the widget subinterface:

../_images/OWArgExplorer1.png