Home // HUSO 2017, The Third International Conference on Human and Social Analytics // View article

Component based Agent Simulation Modeling for Self-Evolving House Market Prediction

Authors:
Joon-Young Jung
Jang Won Bae
Chunhee Lee
Euihyun Paik

Keywords: Agent Simulation; Component; Self-Evolvement

Abstract:
Agent-based microsimulation is used to predict recent social problems because actual population forecasting is expensive and impossible. An agent-based model (ABM) models agents and each individual's interaction among agents. The behavior of agents is mostly defined according to the rules in ABM. However, Long-term simulations using rule-based ABMs accumulate simulation errors. To avoid error accumulation, the simulation model must be autonomously reconfigured using the actual data during the simulation. In this paper, we propose a component based agent simulation modeling for model reconfiguration autonomously. To evaluate the effectiveness of the component based agent simulation modeling for self-evolving simulation, we implement house market ABM simulation system using discrete event system specification (DEVS) C++ engine. As the iteration is progresses, the house market ABM is modified autonomously and the difference between validation data and simulation result is reduced.

Pages: 35 to 38

Copyright: Copyright (c) IARIA, 2017

Publication date: July 23, 2017

Published in: conference

ISSN: 2519-8351

ISBN: 978-1-61208-578-4

Location: Nice, France

Dates: from July 23, 2017 to July 27, 2017