You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Defining mappings across semantic spaces is a hazy business, because the conceptual models underlying the semantic spaces we are mapping are rarely explicitly defined. Even when mapping between two resources that model entities from the same category (disease, gene, phenotype, chemical entity), the underlying conceptual models can differ quite a bit, so there is always some level of conflation (even if what ICD 10 calls a disease is something else then what Mondo defined as a disease, we still want to map ICD10:Diabetes to MONDO:Diabetes for practical reasons). It gets worse though when we have some resources distinguishing certain categories cleanly (say, disease from disease susceptibility, or disease from clinical finding, or disease from disorder) and another that conflates them into the same hierarchy (e.g. diabetes is a "clinical finding/phenotype" and not a "disease").
Some people will just conflate, and not document, no matter what we preach. I would prefer to just provide a way to somehow say: in this mapping, we conflate the two (i.e. we use skos:exactMatch where we should have used skos:relatedMatch).
The text was updated successfully, but these errors were encountered:
Defining mappings across semantic spaces is a hazy business, because the conceptual models underlying the semantic spaces we are mapping are rarely explicitly defined. Even when mapping between two resources that model entities from the same category (disease, gene, phenotype, chemical entity), the underlying conceptual models can differ quite a bit, so there is always some level of conflation (even if what ICD 10 calls a disease is something else then what Mondo defined as a disease, we still want to map ICD10:Diabetes to MONDO:Diabetes for practical reasons). It gets worse though when we have some resources distinguishing certain categories cleanly (say, disease from disease susceptibility, or disease from clinical finding, or disease from disorder) and another that conflates them into the same hierarchy (e.g. diabetes is a "clinical finding/phenotype" and not a "disease").
Some people will just conflate, and not document, no matter what we preach. I would prefer to just provide a way to somehow say: in this mapping, we conflate the two (i.e. we use skos:exactMatch where we should have used skos:relatedMatch).
The text was updated successfully, but these errors were encountered: