A Survey of Fuzzy Approaches in Spatial Data Science

Abstract

Spatial data science emerges as an important subclass of data science and focuses on extracting meaningful information and knowledge from spatial data to enable effective communication and interpretation of both spatial data and analytic results. It emphasizes the importance of location and spatial interaction by storing, analyzing, retrieving, and visualizing spatial and geometric information. Frequently, spatial objects are afflicted by spatial fuzziness, characterizing spatial objects with blurred interiors, uncertain boundaries, and imprecise locations. Fuzzy set theory and fuzzy logic have become powerful tools to adequately represent spatial fuzziness. This paper provides a survey and a review of the literature to understand the application of fuzzy approaches to spatial data science (projects) with the objective of proposing, motivating, and envisioning fuzzy spatial data science.