Publication Type:
Conference Paper
Source:
Workshop W31 Social Sumulation of the 21st International Joint Conference on Artificial Intelligence, Pasadena, California, USA, p.75-88 (2009)
Keywords:
agent-based modelling,
clustering,
Data mining,
data-driven,
social simulation
Abstract:
Along the last years, the Data-driven approach in Social Sim-
ulation is gaining more and more strength. Thus, the use of large col-
lections of empirical data is becoming more frequent, and new demands
emerge. This work attempts to contribute to the debate proposing the
intense use of Data Mining for the improvement and development of
Data-driven Agent-Based Models. Therefore, a methodological approach
explaining why and when to use Data Mining is presented, with a formal
description of each stage. Besides, a case study following the proposed
approach is explained, showing its application step by step.
English
French
Spanish