Estimating agents´ attributes using simulated annealing from statistics to realize social awareness

Social awareness is defined as understanding others´ feeling or their environments. Social simulation can be an effective tool for it because it enables us to see results of what we do in our community. In those simulations, information on citizens and their environments should be realistic and concrete, however, models employed by many applications are too simple, and they can only give abstract lessons from their simulation results. In this paper, we develop a heuristic approach using a simulated annealing and compare two objective functions for reconstructing attributes of citizens in social simulation. We employ the real data such as governmental statistics to reconstruct attributes of citizens. Two objective functions are considered to reconstruct them and show demographic pyramids that are reconstructed.