Observations, measurements and semantic reference spaces

Abstract

What is needed to enable communication about observation and measurement results in information systems? Information system ontologies make a certain conceptualization explicit and partially account for the meanings of symbols associated with this conceptualization. Yet, the meaning of signs denoting measurement results such as “10 m”, “red” or “high” cannot be specified with currently available ontologies. They fail to separate the ontological nature of some observable quality from the specification of how to observe and name the measurement result. We employ the foundational ontology DOLCE for characterizing the ontological nature of observable magnitudes. This involves dealing with ontological questions like “What kinds of observable qualities exist, in which entity does the observed quality inhere and how are the magnitudes of the observed quality structured?”. Then, in order to capture the semantic aspects of an observation result, we introduce semantic reference spaces, which help deal with semantic questions like “Do the signs “10 m”, “33 feet” or “shallow” have the same meaning? Do these signs refer to the same entity, e.g. the depth magnitude of a lake? How to establish a unit of measure?. We posit that the semantic questions can be approached efficiently only if agreement is reached on the ontological questions, and show that the specification of the meaning of signs denoting measurement results is enabled via the extension of the foundational ontology DOLCE with semantic reference spaces. This work was conducted while the author (Probst) was working at the Institute for Geoinformatics, University of Munster, Germany.