On the (limited) difference between feature and geometric semantic similarity models

Abstract

Semantic similarity assessment is central to many geographic information analysis tasks. A reader of the geographic information science literature on semantic similarity assessment processes could easily get the impression that two of the most common approaches, the feature model and the geometric model, are incompatible and radically different. Through a review of literature I seek to elaborate on and clarify that these two approaches are in fact compatible, and I finish with a brief discussion of the handling of uncertain and missing values in these representations. © 2011 Springer-Verlag.