Reiter

Visualisation & Networks / Gephi
Gephi is an open source network analysis and visualisation software that enables users to explore and understand complex relationships and patterns in networks. This application is specifically designed for analysing graphs and networks, with a focus on visualising and interpreting the structure, connections and properties of networks.Gephi's advantages over other network analysis tools lie in its ability to create high-quality visual representations while maintaining transparency over the visualisation algorithms. Gephi is based on Java, so the installation of Java is a prerequisite for using the tool.
Java installation
  1. Visit the official Java website.
  2. Click on the "Download Java" button and follow the instructions to download the installation programme.
  3. Start the installation programme and follow the instructions to install Java on your system.
Installation Gephi
  1. Visit the official Gephi website.
  2. Navigate to the "Download" section and download the latest version of Gephi for Windows.
  3. Once the download is complete, extract the archive and start the application.
Java installation
  1. Visit the official Java website.
  2. Click on the "Download Java" button and follow the instructions to download the installation programme.
  3. Open the downloaded installer and follow the instructions to install Java on your Mac.
 Installing Gephi
  1. Visit the official Gephi website.
  2. Navigate to the "Download" section and download the latest version of Gephi for Mac.
  3. Once the download is complete, open the downloaded disc image (.dmg) and drag the Gephi application into your Applications folder.
Installation Java

sudo apt-get update
sudo apt-get install default-jre
Installation Gephi
  1. Visit the official Gephi website.
  2. Navigate to the "Download" section and download the latest version of Gephi for Linux.
  3. Extract the downloaded archive and navigate to the extracted folder.
  4. Open a terminal in the folder and run the Gephi application, e.g. with the command ./bin/gephi.
  1. Visit the official Gephi website.
  2. Navigate to the "Download" section and download the latest version of Gephi for Linux.
  3. Extract the downloaded archive and navigate to the extracted folder.
  4. Open a terminal in the folder and run the Gephi application, e.g. with the command ./bin/gephi.
Visualisation & Networks / Gephi / Data import
In Gephi, you can import different graph data formats such as XML, TTL and Json. For the sake of simplicity, we will use a csv file in the task. This is a data format that consists of comma-separated values. To describe a graph, we need two csv files, one for the nodes and one for the edges. You have already heard about nodes and edges. We will repeat the principles of graph theory here again in short form with practical relevance.
A network consists of nodes and edges. If we represent the sentence "Sherlock teaches Watson" as a network, Sherlock and Watson would be nodes and teaches an edge. If we were to represent this in a csv table, it would look like this:
Here we provide you with a csv file containing the network example from the lecture. 
When you open Gephi, the following window should open automatically in which you can simply select the Open Graph File option.
Alternatively, you can also access the same option in the top menu bar via File >> Open. Here you select the downloaded nodes.csv as import. You should then see the following window.
To ensure that the table is imported correctly, make sure that Comma is selected in the Separator field. Then click Next and Finish. The setting for temporal display is irrelevant for this network. Always make sure to manually select the option „Append to existing workspace“ when importing. As you import nodes and edges individually, you will otherwise have a network with only nodes and one with only edges, which is not expedient. Now also carry out the import steps with the Kanten.csv. If you have carried out the import correctly and in the same workspace, you should now see a network in the overview view.

Gephi user interface

 Gephi has three different views: The Overview, the Data Laboratory and the Preview. 

Overview: 

In the Overview, you can calculate network dimensions and adjust the visualisation of the network based on these network dimensions. In Gephi, the individual windows of your view are called Groups. You can add further views to the windows by right-clicking on the grey bar above the groups. 
At the top left in Apperence, you can set the colour, size and text layout in your network. Your options are still limited here before the network dimensions are calculated.
At the bottom left you will find the Layout group. Here you can select algorithms according to which your nodes and edges are positioned on the graph view. Once you have selected an algorithm, you can start it and observe what happens to your network. Some algorithms stabilise the network after a short time, others can continue to run forever.
At the top right you can see the most important information about your graph, such as the number of nodes, edges and whether it is a directed or undirected graph.
In the centre right you will find Filters and Statistics group. You can show or hide nodes and edges by filter. You also have more options here if your dimensions have been calculated. You can perform the calculation by selecting the Statistics tab here.
At the bottom right you will find the Queries area where you can link filters hierarchically, which is a very advanced option.   

Data Laboratory:

 In the data laboratory you will find your node and edge tables. You can delete, add, change etc. data here. Please also note that all filters that you use in the Overview also affect your data table. In addition, the csv files that you used for the import are not affected by your changes in the data lab, as they are saved in a separate gephi file after the import. 

Preview: 

In the preview, you can make final changes to the display of the network in the export. Whenever you make a change, you must click the Refresh button at the bottom left to regenerate the display.

Task

Customize your imported network and upload your version to the forum. If you want more data to work with, here is a network of  seventeenth-century Quakers (a religious movement), provided by Melanie Walsh: