Comparison of two fuzzy algorithms in geodemographic segmentation analysis: The Fuzzy C-Means and Gustafson–Kessel methods

Clustering techniques are frequently used to analyze census data and obtain meaningful large-scale groups. Geodemographic segmentation involves classifying small geographic areas – for example, block groups, census tracts, or neighborhoods - into relatively homogeneous segments. Most studies concerning geodemographic analysis and fuzzy logic employ the Fuzzy C-Means algorithm. In this paper, we compare two algorithms for fuzzy clustering in geodemographic analysis, and their structures, as well as their pros and cons, are analyzed. These are the Fuzzy C-Means algorithm and the Gustafson–Kessel algorithm The main objective of this paper is to evaluate the performance of the Fuzzy C-Means and Gustafson–Kessel algorithms in the clustering problem, under specific conditions. An experimental approach to this problem is adopted through the use of a real-world dataset describing 52 attributes of the 285 postal codes in the Athens metropolitan area.

► We compare two algorithms for fuzzy clustering in geodemographic analysis.
► We evaluate the performance of the Fuzzy C-Means and Gustafson–Kessel algorithms.
► We present a case study through the use of a real-world dataset.
► We highlight the pros and cons of these algorithms in geodemographics.