Applying Image Texture Measures to Describe Segregation in Agent-Based Modeling
| Autor: | Kathleen Pérez-López |
|---|---|
| Rok vydání: | 2010 |
| Předmět: | |
| Zdroj: | Simulating Interacting Agents and Social Phenomena ISBN: 9784431997801 WCSS |
| DOI: | 10.1007/978-4-431-99781-8_16 |
| Popis: | Many agent-based models produce output as multi-region patterns on a two dimensional grid. The degree of mixing versus segregation in these displays is a particularly interesting outcome of some of these models, and comparing output patterns is an important component of evaluating their results. Typically these comparisons are made in an ad-hoc and qualitative manner. This paper proposes applying measures used to describe textures in the field of image processing to quantitatively evaluate patterns resulting from agent-based models. The rationale for using these measures is made by relating the underlying image formation process, based on Markov Chain Monte Carlo methods, to the Schelling segregation process. Examples are presented that support the premise that these statistics capture the degree of segregation in a 2D display. The ideas are also tested on the output of the agent based model, Wetlands (Cioffi-Revilla C, Paus S, Luke S, Olds JL, Thomas J (2004) Mnemonic structure and sociality: a computational agent-based model. In: Conference on collective intentionality IV, Certosa di Pontignano, Siena, Italy, 13–15 October 2004), supporting the premise. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|