Representing dispositions

Johannes Röhl,Ludger Jansen

doi:10.1186/2041-1480-2-s4-s4

Abstract

Dispositions and tendencies feature significantly in the biomedical domain and therefore in representations of knowledge of that domain. They are not only important for specific applications like an infectious disease ontology, but also as part of a general strategy for modelling knowledge about molecular interactions. But the task of representing dispositions in some formal ontological systems is fraught with several problems, which are partly due to the fact that Description Logics can only deal well with binary relations. The paper will discuss some of the results of the philosophical debate about dispositions, in order to see whether the formal relations needed to represent dispositions can be broken down to binary relations. Finally, we will discuss problems arising from the possibility of the absence of realizations, of multi-track or multi-trigger dispositions and offer suggestions on how to deal with them.

Highlights

Terms for dispositions and their cognates like tendencies or propensities are important in biomedical data structures [1]
The task of representing dispositions in some formal ontological systems is fraught with several problems, which are partly due to the fact that Description Logics can only deal well with binary relations
The paper will discuss some of the results of the philosophical debate about dispositions, in order to see whether the formal relations needed to represent dispositions can be broken down to binary relations

Summary

Introduction

Terms for dispositions and their cognates like tendencies or propensities are important in biomedical data structures [1]. Dispositions relevant for the biomedical domain comprise the tendency of a patient for nausea, the capacity of drugs like aspirin to relieve pain, and the dispositions of molecules to undergo certain chemical reactions under certain circumstances Such dispositions are important, because in the medical domain we are in general interested in what happens, and in what could happen, given the causal properties of the patient, as we may need this knowledge, e.g., in order to prevent something from happening by taking precautions, and disposition ascriptions entail which occurrences could take place if certain conditions were met. We will briefly discuss two applications of dispositions in the ontology of biology and medicine that show why these features make dispositions crucial in many respects

Methods

Results

Conclusion