An agent-based simulator driven by variants of Self-Organizing Maps

摘要: Abstract This paper introduces an agent-based simulator driven by variants of Self-Organizing Maps (SOMs), specifically designed to model agents learning in economic systems, as well render how they interact and the way such interaction can affect system general behavior. As a consequence, we developed environment with SOMs nodes treated that are suitable simulate systems their evolution over time; moreover, this were able study within SOM framework impact spatial connections on individual decisions. The effectiveness has been tested formalization growth. Agents behavior is simulated when production efforts direct consequence individuals (in our simulation: nodes) allocate time energies between working studying, thus defining corresponding consumption savings patterns. We then coherence respect observable data. results confirm that, order dynamics, it relatively easy mold so simulation highlights significant Furthermore, examined case being patterns consistent existence dichotomous growth, i.e. combination convergence regions divergence among regions, be help rulers effectively address policy intervention.