Exploring and understanding outputs from agent-based models is challenging due to a relatively higher number of parameters and multidimensionality of the generated data. We use a combination of exploratory and data mining techniques to understand data generated from an existing agent-based model to understand the model’s behavior and its sensitivity to initial configurations of its parameters. This is a step in the direction of an ongoing research in the social simulation community to incorporate more sophisticated techniques to better understand how different parameters and internal processes influence outcomes of agent-based models.