Modeling and Analysis
      Back to home
      1. Innoslate Help Center
      2. Walkthrough Series
      3. Modeling and Analysis

      Social Network Analysis Walkthrough

      Understand how to run a Social Network Analysis (SNA) in your Innoslate models.

      Understanding Social Network Analysis Centrality Measures in Innoslate

      Innoslate utilizes key concepts borrowed from Social Network Analysis (SNA) to help you understand the relationships within your network. In this guide, we’ll break down these concepts using Innoslate-specific terminology and provide a step-by-step walkthrough of how to measure and calculate different centrality scores for nodes (assets) in a network.

      ⚠️It is important to note that the SNA report is currently only available in Innoslate v4.11 or above.

      * To follow along and try out the features mentioned, click here to download the Social Network Analysis Walkthrough Project .INNO File and import it into a new Innoslate project.

      SNA Terminology in Innoslate

      Before we dive into the centrality measures, let’s clarify the terms we’ll be using throughout this guide:

      • Nodes = Assets
        These are the individual entities in your network.

      • Connections = Conduits
        These represent the relationships or links between the assets.

      • Network map = Asset Diagram
        This is the visual representation of the network of assets and their connections.

      Throughout this guide, we'll be referring to Figure 1 (below) as our example to explain the concepts.

      2441_tile

      Figure 1: Example Network Map

      In this figure, nodes (or assets) are represented by squares, and the conduits (or connections) between them are shown as lines linking those circles. As we discuss each centrality measure, we'll refer back to this example. 

      If you downloaded the Social Network Analysis Walkthrough, you'll be focused on the Social Network Analysis Asset Diagram.

      1. Degree Centrality: Counting Direct Connections

      Degree Centrality is one of the simplest centrality measures, but it provides valuable insights into how "central" or "well-connected" a node is in a network. In Innoslate terms, Degree Centrality counts how many "connected by" relationships an asset has.

      • How to calculate Degree Centrality:
        • Simply count the number of conduits (connections) linked to an asset.
        If an asset has three conduits connected to it, its Degree Centrality score is 3. This indicates that the asset is directly linked to three other assets in the network.

      2. Closeness Centrality: Measuring Proximity to Other Nodes

      Closeness Centrality measures how close a node is to every other node in the network. The closer an asset is to all others, the higher its Closeness Centrality score. This measure reflects how quickly a node can reach every other node in the network, based on the shortest path between them.

      How to Calculate Closeness Centrality:

      Follow these steps:

      1. Determine the shortest path between each pair of assets. The shortest path is the one that uses the least number of conduits.
      2. Sum the lengths of all the shortest paths from your node to every other node. For example, if Node #1 is connected to Node #2 with 2 conduits and to Node #3 with 3 conduits, the sum would be 5 conduits.
      3. Normalize the score by dividing the number of nodes minus one (i.e., the number of nodes excluding the one you’re measuring) by the sum of the shortest paths.

      Example:

      • Suppose you have 6 assets in the network, and for Node #1, the sum of the shortest paths is 7.
      • The normalized score would be calculated as follows:
        (6−1)/7=5/7=0.714(6 - 1) / 7 = 5 / 7 = 0.714(6−1)/7=5/7=0.714

      This gives you a sense of how quickly Node #1 can access the other assets.

      3. Betweenness Centrality: Identifying Choke Points

      Betweenness Centrality measures how often a node acts as a bridge or choke point in the network, i.e., how often it lies on the shortest path between other pairs of nodes. A high Betweenness Centrality score indicates that the asset plays a crucial role in connecting different parts of the network.

      How to Calculate Betweenness Centrality:

      1. Identify all pairs of nodes in the network and determine how many shortest paths exist between them.
      2. Count how many times each asset lies on these shortest paths.
      3. Divide the count by the total number of shortest paths between the nodes in question.
      4. Sum the values from all pairs to get the Betweenness Centrality score.

      For instance, in Figure 1, Node #2 is a key asset because it lies on the shortest paths between Nodes #3 and #4. This makes Node #2 a bottleneck in the flow of information.

      4. Eigenvector Centrality: Quality of Connections

      Eigenvector Centrality is a more advanced measure that looks beyond the quantity of connections and evaluates the quality of those connections. It considers how connected your neighbors are and gives higher scores to nodes that are connected to other important nodes.

      How to Calculate Eigenvector Centrality:

      1. Create an adjacency matrix for your asset diagram. This matrix represents the connections between each pair of assets.
      2. Find the largest eigenvalue of the adjacency matrix. This eigenvalue represents the importance of the connections in the network.
      3. Identify the corresponding eigenvector to the largest eigenvalue. This vector indicates the relative importance of each node.
      4. Normalize the eigenvector so that the largest entry equals 1. This gives you the final Eigenvector Centrality score.

      In practical terms, assets with high Eigenvector Centrality are highly connected to other well-connected assets, which often means they play a pivotal role in the overall structure of the network.

      5. How to run a Social Network Analysis Report

      1. Navigate to the specific Asset Diagrams that include the required centralities mentioned in the above prerequisites.
      2. Find and Select 'Report' on the toolbar.
      3. Select 'Social Network Analysis.' The CSV download will automatically begin.

      Figure 2: Figure 1's Social Network Analysis Report (CSV) Example

      Wrapping Up: Using Centrality Measures in Innoslate

      Understanding the different types of centrality in a network helps you assess the roles and influence of various assets within a system. Whether you’re managing an Asset Diagram in Innoslate for system optimization or risk analysis, centrality measures can provide valuable insights into how information flows through your network and highlight key assets that might require attention. By adhering to the procedures detailed in this guide, you will be able to compute Degree, Closeness, Betweenness, and Eigenvector Centrality for any network. You can then generate reports and leverage these insights to inform data-driven decisions within Innoslate.