Complex networks involve objects in a system, how they relate to one another and how they affect the behavior of the system as a whole. Because the network may be large and the relationships complicated and non-linear, traditional analytical tools cannot capture or predict system behavior.
Complexity derives from both the objects (nodes) and the links between them. Nodes can be anything from people to valves, cities to molecules; they are basically things and are defined by their properties or characteristics.
Complexity derives from the links between nodes, which are more multifaceted than the nodes. They can be bi-directional, distributed in various patterns, and can vary in importance (weight). For instance, in a network of people, a weight could be assigned to the likelihood that two people will interact. In a computer system, the links might represent bandwidth.
It is the distribution of the links and distribution of the weights that analysts look at; these are what determine the behavior of the network, regardless of what the objects or nodes are.
What are the key benefits?
Analyzing complex networks provides insight into what causes change in the behavior of the system.
- Network analysis shines light into the inner workings of the network, exposing the underlying reasons for specific network behaviors that one could not otherwise isolate.
- One of the most important benefits of network analysis is the ability to identify patterns and understand how to influence them.
Some possible applications
- Placing advertising to take advantage of a known consumer network.
- Understanding how a specific social network operates and how it will respond to changing behavior of its members.
- Identifying the best places in a pipe system to install safety valves that will provide necessary protection without compromising day-to-day flow or operation.