An Ontology Design Pattern to Represent False-Results

Observations are an important aspect of our society. Arguably, great part of them is captured by means of sensors. Despite the importance of the matter, the ontology of observations and sensors is not well developed, with few efforts dealing with the fundamental questions about their nature. As a result, an important aspect of sensors is overlooked: sensors may fail, producing false-results (i.e. false-positives and false-negatives). The lack of a proper representation of this aspect prevents us from communicating and reasoning about sensor failures, making it harder to assess the correctness of observations and to treat possible errors. In view of this problem, we propose an ontology design pattern (ODP) to represent false-results of sensors. It covers a special case of sensor that exclusively produces positive or negative results regarding the presence of the type of entity the sensor is designed to perceive. The paper introduces the ODP structure as well as its ontological commitments, bringing an example from the biomedical field. Discussion and further research opportunities of research are posed at the end of the paper.