Fuzzy Bayesian inference for mapping vague and place-based regions: a case study of sectarian territory

The problem of mapping regions with socially-derived boundaries has been a topic discussion in the GIS literature for many years. Fuzzy approaches have frequently suggested as solutions, but none adopted. This is likely due to difficulties associated determining suitable membership functions, which are often arbitrary crisp that they seek replace. paper presents novel approach fuzzy geographical modelling replaces function possibility distribution estimated using Bayesian inference. In this method, data from multiple sources combined estimate degree given location member set and level uncertainty estimate. Inference demonstrated through case study census perceptual behavioural evidence model territory two segregated groups (Catholics Protestants) Belfast, Northern Ireland, UK. method provides robust empirical basis use models GIS, therefore applications range otherwise vague boundaries.