preview

Using Multi Agent System ( Mas )

Better Essays
Near Term Analysis: To generate “what-if” scenario a multi-agent model called “Near term Analysis” introduced. This framework generates “what-if” scenarios by processing the input dataset with the social network analysis method. To perform “what-if” analysis of an organization under different possible threat scenarios are done by using Multi-Agent system (MAS) called Dynet. This framework puts Dynet in data framing environment so that large number of simulations can be run with different possible threat scenarios.
Working of Near Term Analysis: The Near term analysis frame work takes meta-matrix as input and then the agent interaction mechanism takes place. The isolating of agents is the threat scenario that is
…show more content…
Meta-matrix contains various kinds of nodes and internode type links. This network has sub-network such as agent-agent network, agent- knowledge network. By including these networks the interactions among the agents can be simulated. Illustrative example of Meta-matrix network is shown below Agents Knowledge Tasks
Agents Social Network Knowledge Network Assignment Network
Knowledge Information Network Needs Network
Tasks Precedence Ordering

Agent-Knowledge Network: The knowledge network is “who knows what” in organization. Knowledge is defined as different categories that are relevant to particular organization. For example, if we are collecting the data about organizational simulation group we may have the information like software development, organizational theory and statistics. The knowledge network is simply who possess what level of expertise in that particular field.

k1 k2 k3 k4
A1 1 0 0 1
A2 1 1 0 1
A3 1 0 0 1
A4 0 1 1 0 An Illustrative example for Agent-knowledge matrix .
“0” indicates that particular agent has no knowledge about the particular knowledge bit and “1” indicates otherwise. From above table we can say that A1 has knowledge about knowledge bit k1 and similarly A2 has no knowledge about the knowledge string k3.
Agent-interaction mechanism: The agents in this model have the opportunity to interact with others
Get Access