Initializing Agent-Based Models With Clustering Archetypes

Agent-based models are a powerful tool for predicting population level behaviors; however their performance can be sensitive to the initial simulation conditions. This paper introduces a procedure for leveraging large datasets to initialize agent-based simulations in which the population is abstracted into a set of archetypes. We show that these archetypes can be discovered using clustering and evaluate the benefits of selecting clusters based on their stability over time. Our experiments on the GitHub dataset demonstrate that simulation runs performed with the clustering archetypes are more successful at predicting large-scale activity patterns