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
This issue is to strongly recommend that we provide a brief definition of the ROBOKOP scoring/ranking algorithm to users by way of hovering over the "score" column in answer sets or perhaps providing a legend somewhere on the question-builder.
Here's a draft definition (sans references):
The ROBOKOP scoring/ranking algorithm employs formalism developed within electrical resistance theory. In brief, the algorithm finds the highest scoring path between a pair of nodes of interest, and the scores are based primarily on publication support for edges. These edges are either derived from curated knowledge sources and/or contributed by a ROBOKOP service termed OmniCorp, which exposes entity co-occurrence terms derived from PubMed abstracts; the former are weighted more heavily than the latter. The scores that are assigned to specific answer subgraphs within a result set should be considered relative to one another and not absolute or comparable across result sets.
The text was updated successfully, but these errors were encountered:
This issue is to strongly recommend that we provide a brief definition of the ROBOKOP scoring/ranking algorithm to users by way of hovering over the "score" column in answer sets or perhaps providing a legend somewhere on the question-builder.
Here's a draft definition (sans references):
The ROBOKOP scoring/ranking algorithm employs formalism developed within electrical resistance theory. In brief, the algorithm finds the highest scoring path between a pair of nodes of interest, and the scores are based primarily on publication support for edges. These edges are either derived from curated knowledge sources and/or contributed by a ROBOKOP service termed OmniCorp, which exposes entity co-occurrence terms derived from PubMed abstracts; the former are weighted more heavily than the latter. The scores that are assigned to specific answer subgraphs within a result set should be considered relative to one another and not absolute or comparable across result sets.
The text was updated successfully, but these errors were encountered: