1 / 11

How much formality do we need ?

MedInfo 2007 Workshop: MedSemWeb 2007 What Semantics Do We Need for A Semantic Web for Medicine?. How much formality do we need ?. Stefan Schulz. University Medical Center Freiburg, Medical Informatics, Freiburg, Germany. Example. Using Semantic Web standards (OWL-DL)

nevaeh
Download Presentation

How much formality do we need ?

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. MedInfo 2007 Workshop: MedSemWeb 2007 What Semantics Do We Need for A Semantic Web for Medicine? How much formality do we need ? Stefan Schulz University Medical Center Freiburg, Medical Informatics, Freiburg, Germany

  2. Example • Using Semantic Web standards (OWL-DL) • Using Biomedical Ontology standards (OBO) • Terminological Inference

  3. Classes • Amino Acid • Protein • Aminoaciduria • Proteinuria

  4. Relations (OBO RO) • hasPart / partOf(parthood in a broad sense):relates continuants • hasLocation / locationOfrelates continuants or occurrents with continuants • transitive, reflexive, antisymmetric

  5. Description Logic • Subsumption ⊑ • Equivalence ≡ • Existential quantification  • Conjunction ⊓ • transitive roles

  6. Axioms • Protein⊑hasPart.AminoAcid • Aminoaciduria≡Disorder ⊓ • hasLocation.(Body ⊓ • hasPart.(PortionOfUrine ⊓ • hasPart.AminoAcid)) • Proteinuria≡Disorder ⊓ • hasLocation.(Body ⊓ • hasPart.(PortionOfUrine ⊓ • hasPart.Protein))

  7. false! Inference Proteinuria⊑Aminoaciduria (since Proteins have Amino Acids as parts, and partOf is transitive) • Is this error due to formal underspecification ? • Is hasPart not always transitive?

  8. Formal correctnessbut ontological sloppyness AminoAcid: hidden ambiguity: • AminoAcidSingleMolecule • AminoAcidResidue • AminoAcidSingleMoleculeCollection • AminoAcidSingleMoleculeCollectionLowConc • AminoAcidSingleMoleculeCollectionHighConc

  9. Corrected Axioms • Aminoaciduria≡Disorder ⊓ • hasLocation.(Body ⊓ • hasPart.(PortionOfUrine ⊓ • hasPart.AminoAcidSingleMoleculeCollectionHighConc)) • Proteinuria≡Disorder ⊓ • hasLocation.(Body ⊓ • hasPart.(PortionOfUrine ⊓ • hasPart.ProteinMoleculeCollectionHighConc))

  10. Formal Correctness assures consistency Ontological Correctness assures adequacy Two sides of the same coin

  11. Conclusion • Even little formality must be rooted in a correct ontological foundation to prevent unintended models with inadequate inferences • If we do not know exactly what we are formalizing we cannot rely on machine reasoning. In this case we should give preference to informal, thesaurus-like knowledge representations

More Related