BROWSE
Detail
Conference Paper
Component Based Agent Simulation Modeling for Self-Evolving House Market Prediction
Cited - time in 
Share 
Share
- Authors
- Joonyoung Jung, Jang Won Bae, Chunhee Lee, Euihyun Paik
- Issue Date
- 2017-07
- Citation
- International Conference on Human and Social Analytics (HUSO) 2017, pp.35-38
- Language
- English
- Type
- Conference Paper
- Abstract
- Because actual population forecasting is expensive and impossible, recent agent-based microsimulation is used to predict social problems. An agent-based model (ABM) models the interaction between an agent and each agent. However, long-term simulations using rule-based ABMs accumulate simulation errors. To avoid this error accumulation, the simulation model must be dynamically reconfigured using the actual data during the simulation. In this paper, we propose a component-based agent simulation modeling for model reconstruction and implemented a housing market ABM simulation system using DEVS (Discrete Event System Specification) C++ engine to evaluate the effect of error accumulation avoidance. As the simulation progresses, the housing market ABM is dynamically modified to reduce errors between the actual data and the simulation results.
- KSP Keywords
- Agent simulation modeling, Component-based, Discrete Event(DE), Error accumulation, Housing Market, Long-term simulation, Market prediction, Rule-based, Simulation system, Social problems, agent-based model(ABM)