Social Network Visualizer  2.2 Social Network Visualization and Analysis Software
User Interface overview

SocNetV has a simple Graphical User Interface (GUI) composed of:

At the top of the window, there is the menu bar, filled with commands and options, organized in 6 menus:

• Network: Options to load and save a network, export it, create random nets etc.
• Edit: Options to add/remove nodes and edges, change colors, filter edges/isolate nodes, etc.
• Layout: Options to apply layout methods, i.e. reposition all nodes according to their centrality.
• Analysis: Gives you tools to analyse the active network (density, diameter, centralities, distance matrix, etc).
• Options:Allows you to show/hide edges, edge arrows, turn on/off antialiasing, etc.

# The toolbar

Below the menu, the toolbar lets you carry out a range of actions, such as: new network creation, load a network file, save current network, print as well as switch between relations, add new relations and display help messages for the menu options.

The main part of the application window is occupied by the sidebar panels (left-right) and a virtual "canvas" (right-side) where network nodes and edges appear.

# The Panels

The panel on the left of the window is the Control Panel. The panel on the right is the Statistics Panel.

## Control Panel

At the top of the Control Panel there are 4 buttons to edit the network (add/remove node, add/remove edge).

Below them, there are two groups of options to Analyze and Visualize the loaded network.

In the Analyze group, options are categorized in four submenus:

• Distances: Options to compute distances, geodesics, eccentricity and diameter
• Connectivity: Options to compute connectivity, walks and reachability matrix.
• Clusterability: Options to compute cliques, triad census, clustering coefficient etc.
• Prominence: Options to compute one of the many Centrality and Prestige indices that SocNetV supports.

When you select an option, SocNetV computes what you asked and displays the report in a new texteditor window.

Note
All reports are automatically saved in a directory called "socnetv-data" in your HOME folder. Of course, you can edit the report as you wish and save it to another filename.

In the Visualize group, there are menus and checkboxes to embed visualization layouts to the current network.

With one click, SocNetV can visualize the network in some intuitive ways. There are two layout categories:

• By prominence indices. Here you can select a prominence metric (i.e. Betweenness) and a layout type (i.e. circular). Hit "Apply" and voila!
• By dynamic models (i.e. Force directed), such as the Eades model.

Finally, at the end of the toolbox there are options to toggle node sizes according to their inDegree/outDegree and enable/disable layout guidelines.

## Statistics Panel

The Statistics Panel is mainly occupied by informative LCDs.

At the top of the Panel there is a Network Type label, which indicates whether your data is directed or undirected.

The LCDs below display statistics for the active network (i.e. node and edge counts, density, counts of inLinked/outLinked nodes, etc). Furthermore, under "Active Node" there are LCDs which refer to the last clicked node (notably its number, in-Degree, out-Degree and Clustering Coefficient).

# The canvas

The canvas is the main area of interaction. You can:

• point and click on a node/edge to select it, left-click to open context menu
• double-click on empty space to add a node,
• middle-click on a node to add a directed edge from it.

The initial background color is set to "white", but you can changed it from the Edit -> Colors menu.

Below, we describe how to work with SocNetV.

# Working with SocNetV

## Network creation

To start working with SocNetV you need network data, i.e. a graph of nodes (vertices) and links (edges).

SocNetV enables you to create networks by point and clicking on the canvas or load them from files.

There are multiple ways to create or edit nodes and links in SocNetV:

• from the left panels buttons, or
• by right/left/middle/double-clicking on the canvas.

## Creating and handling nodes

To create a new node, you can double-click on the canvas, click on the "Add node" button or press CTRL+.

You can move a node by left-clicking on it and dragging it with mouse.

Right-click on a node to display a context menu with options to delete it, add edge, change node properties (see below). To change the color, size or label of a node, right-click on it and in the context menu select Node Properties.

All nodes by default are tagged by their node number. If you want to display the labels as well, enable the option in the menu Options -> Node -> Display Labels.

Warning
In large networks, it is sometimes difficult to locate a specific node. In such cases, you can press Ctrl+F to find a desired node (by nodenumber or label). SocNetV will highlight that node for you. Press Ctrl+F again to undo this.

When you right-click on a node, a context menu appears. From there you can remove the node, add an edge or select Node Properties to change its color, label, size, shape, etc. A similar menu appears when you right click on an edge.

From the Node Properties dialog, you can enter a node label, change the node size (and "value" in future versions), edit the node color and select a proper node shape. SocNetV supports many kinds of node shapes, i.e rectangles, diamond, ellipse, circle, etc. Click OK and all your changes will be done in one step. For instance clicking on the red color button will bring up a nice Colors dialog where you can select every possible color.

Group select If you want to select more than one node, press and hold down the left mouse button on the canvas. By moving your mouse, a rectangle will be drawed. All nodes inside this rectangle will be selected the moment you release the mouse button. You can also edit multiple selected nodes at once. Left-click on the canvas and drag to make a selection area. Then right click on one of the selected nodes to open the Node Properties dialog for all of them. Any change you make will be applied to all selected nodes.

Warning
In networks with thousands of edges, the group selection process can be slow...

## Creating and editing edges

There are lots of ways to create edges in SocNetV.

For instance, you can do it by middle-clicking or double-clicking on a source node and then middle-clicking or double-clicking again on a target node.

By default, all new links created this way have a unit weight (w=1). You can change the weight of an edge later by right-clicking on it and selecting "Change Weight" from the menu which appears.

Note
By default, edges are drawn so that their line width is analogous to their weight. The formula used is $$width = 1 + \log(|weight|)$$

You can also create edges by right-clicking on any node (thus selected as source), and selecting "Create Edge". In the new dialog, just enter the target node number and the desired edge weight.

Finally, you can click on the "Add edge" button from the left Control Panel. In that case, you will be asked for both the source and the target node numbers (and the edge weight).

If you prefer using the keyboard, you can also create new edges using the keyboard shortcut CTRL+/. You will be asked for source and target nodes and their weight.

Note
When you create the first edge of a network, you will be asked to label the current Relations relationship.

Edge Creation Example:

Say you created two nodes, numbered 1 and 2, on the canvas. To create a new directed edge from node 1 to node 2, middle click on node 1 (the mouse pointer will become a hand) and afterwards middle-click on node 2. A new line will be drawn instantly. If you want an undirected edge (edge) repeat the process from node 2 to node 1.

Remember, each edge you create this way has the default weight 1 and black colour.

Right-click on an edge to display a context menu with options to delete it, change weight, color etc.

Note
When you click on an edge, SocNetV highlights the source and target nodes for your convenience. Click again the edge to undo this.

# Relations

The first time you create an edge in your network, the application asks you to enter a name (or label) for the new relation between actors/nodes.

A relation is a collection of ties of a specific kind between the network actors.

For instance, you might want to label a relation "friendship" if the edges between nodes refer to the set of friendships between pairs of actors.

Starting from version 1.3, SocNetV supports multirelational networks, that is networks with ties of different kind between actors.

You can add more relations to your network by pressing the + button in the toolbar.

You can switch between relations by clicking the previous and next arrow buttons in the toolbar.

Note
While modifying a multirelational network, you can add more nodes but you may not remove a node from the network.
Warning
Each time you save a network, SocNetV saves the active relation only. You can save another relation by moving to it and then save again to another file.

# Basic functions in Network Menu

The easiest way to start working with SocNetV is when you have already network data saved in a supported format (see Supported Formats [Formats].

For instance, you might have another program (for example a simulation) creating adjacency networks which you want to visualise. In that case, from the SocNetV's menu go to File > Load. In the dialogue that will appear, navigate to the desired folder and select the appropriate network file.

Once you select a file, SocNetV will first display it in a File Previewer window. This allows you to select -if you need to- a specific codepage for that file. You might need this feature due to the different codepages used by the Windows, Linux and Mac operating systems. For example, say you use SocNetV in Mac OS X and you want to open a Windows network file containing non-Latin characters (such as é or some curillic "кЧДЛХКЮ"). To properly open that file you need to select its original codepage, i.e. KOI8-R. That's what File Previewer allows you to do.

Of course, if the file has been created in the same OS as yours or by SocNetV you do not need to select another codepage, just press OK to load the network in the canvas.

Note
The default codepage for loading files is UTF-8. Select another codepage only if you know that the file comes from Windows or has non-latin characters.
Linux and Mac users should always use UTF-8, except when they try to load files saved in Windows computers.
Windows users should probably use Windows-1253, except when they want to load files saved in Mac/Linux or files containing non-latin chars (i.e. Russian).
Russian Windows users should probably use KOI8-R encoding, except when reading files saved from SocNetV (UTF-8)

## Saving the active network

To save the active network, just press Ctrl+S or click on the menu entry File > Save. By default, it will be saved in GraphML format.

If you like, you can export it to another supported format (menu Network > Export To). Note that some Supported Formats [Formats] are supported only for loading - not for saving.

Note
Each time you save a network, SocNetV saves the active relation.
By default, UTF-8 is used for writing files. This means SocNetV by default uses UTF-8 codec for all output textstreams, such as network files.

The adjacency matrix of a network is a matrix where each element a(i,j) is equal to the weight of the edge from node i to node j. If the nodes are not connected, then a(i,j)=0.

To view the adjacency matrix of a network, press F6.

By default, SocNetV displays the adjacency matrix as integer-valued only (although we do allow float weights).

## Known data sets

SocNetV can recreate known data sets. Network Generation [Read more].

## Random network creation

SocNetV can create a random network for you. Network Generation [Read more].

## Web Crawler

SocNetV includes a simple web crawler. Network Generation [Read more].

## Printing and Exporting

To print the network directly to your printer, press Ctrl+P.

Keep in mind, that SocNetV follows the "what you see is what you print" principle: we print what is viewable in the canvas, i.e. if you zoom-in to a network, the application will only print that specific network portion. So, you might need to zoom-out enough so that the whole network is viewable and therefore printable.

Except printing, you can export your work into raster (BMP and PNG) images, as well as PDF documents. The latter are vector-based, and therefore offer the best quality. Again, keep in mind the rule "what you see is what you print".