Using the demo version of Network Analysis Engine, you can try some standard network topologies and see the associated entropy spectra analysis. The standard network topologies it produces are

• Ring Structure Network
• Tree Structure Network
• Clustered Network

• Ring Structure

  This is the simple network structure in which each node is connected to its closest neighbor. Each node is connected to the next and the last node is connected to the first one making a completely closed set. Since all of the nodes are connected with two other nodes, all entropy values are same. You can try this network with different numbers of nodes and see the flat entropy graph.

  

• Tree Structure

  The ExaSphere Demo shows two types of tree structures - a binary tree structure and the trinal tree structure. In the binary tree, a node has two child nodes and in the trinal tree has three child nodes. You can select any tree type and the tree depth to see the entropy spectra. In the binary structure diagram, every node is connected to its two children with a weight of 1 and connected back to its parents with a weight of 1. You can see three groups of nodes in the entropy graph, except for the case where the tree depth is 1. When tree depth is 1, there are two different groups. The first group which has the higher value in the entropy graph is the group of nodes on the right side in the diagram. Each of these nodes has only one connection (to its parent) so the entropy values are high. The second group is the starter node. It has two connections (with its direct children) and has the middle entropy value. The third group are the middle nodes. They each have 3 connections (with one parent and two children). This is the most active group in this diagram and thus has the lowest entropy value. The trinal tree works in similar way. You can click the left mouse button on the graph to see which node it belongs to. The node index increase left to right, top to bottom. If you want to give a different weight for a different depth, you can draw your own network diagram on the "User Definable Network" panel.

  

• Clustered Network

  The diagram below shows 3 nodes that are clustered with a weight of 5 and a network size of 10. As you see in the diagram, some nodes are higly connected with other clustered nodes. The clustered nodes are symmetrically connected with other clustered ones with using the user-defined weight. The others are randomly connected with weight 1. Each clustered node shows low entropy. Click the left mouse button on the graph to see which nodes it belongs to. You can try this network with different network sizes and with different clustered situations to see how the entropy varies.

  

• User-Definable Network

  You can draw your own network diagram here. On the left side, you can create a network diagram using a mouse - click the left mouse button to create a node and drag it to draw a network line between two nodes. Each network line has a default network flow weight of 1 in both directions. For example, a line between node "0" and "1" means the network flows with a weight of 1 from node 0 to node 1 as well as from node 0 to node 1. You can edit the network flow by right-clicking on the network line. If you want to give different weights for different directions, select the anti-symmetry option and insert any value. Anti-symmetric network flows appears as a blue line and symmetric flow appears as a black line on the diagram. Thicker lines represent higher weights. When you finish drawing the network diagram, press the "Calculate Entropy" button to see the entropy graph. Click the left mouse button on the graph to see the nodes to which it belongs.

  

In order to run the demo, you need the Java plugin for your browser. The easiest way to get the plugin is to install the Java Runtime Environment (JRE) which you can download for free from here.
Launch the ExaSphere Demo