Fuzzy approaches to context variable in fuzzy geographically weighted clustering

TL;DR

This paper introduces fuzzy methods to determine context variables in FGWC, enhancing computational efficiency and focus in geo-demographic data analysis, with a numerical example demonstrating their application.

Contribution

It proposes two novel fuzzy approaches for setting context variables in FGWC, addressing the challenge of defining crisp values.

Findings

Fuzzy methods improve FGWC computational speed.

Proposed methods effectively focus clustering results.

Numerical example validates the approaches.

Abstract

Fuzzy Geographically Weighted Clustering (FGWC) is considered as a suitable tool for the analysis of geo-demographic data that assists the provision and planning of products and services to local people. Context variables were attached to FGWC in order to accelerate the computing speed of the algorithm and to focus the results on the domain of interests. Nonetheless, the determination of exact, crisp values of the context variable is a hard task. In this paper, we propose two novel methods using fuzzy approaches for that determination. A numerical example is given to illustrate the uses of the proposed methods.