Pattern-Oriented Agent-Based Simulation for Analyzing Family Strategies in Civil Service Examination in Imperial China

摘要

Pattern Oriented Modeling (POM) is an approach to bottom-up complex system analysis which was developed in ecology and for agent-based complex systems. This paper proposes a pattern-oriented agent-based simulation (POABS) approach to analyze agent-based complex system. We apply POABS to a history simulation domain to analyze a particular family line with more successful candidates in the civil service examination in imperial China. In order to decode family strategies along such an elite family line, we develop POABS to test relevant patterns observed in the real family system. Inverse simulation technique is applied to evaluate each simulated pattern through fitting the time-series of simulated profile data to real profile data with real-coded GA. Intensive experiments show a practical applicability of POABS in agent-based complex system.