I've written rules that negate terms by looking for two types of tokens (according to dictionary lookups): Negations (not, no, etc.) and Diseases.
The rule is correctly removing the Disease annotation from the analysis.
I have five medical dictionaries (diseases, procedures, medications, diseases, physical conditions, etc.) so I want to generalize the rule to remove all these annotations.
I tried making a group (requiring 1 occurrence) and with group members as 0 or 1 occurrence.
This time when I tried to right-click one of the annotations listed in the group and chose "Delete Annotation" the annotation actually disappeared (not like above where it is grayed out and listed as a plain token). I'm not sure I understand the interface in this case.
So for the text "no mumps" the first rule worked, but the second group-type did not.
Is my logic incorrect? Is there a recommended way to deal with this. It seems that generalizing Annotations in a group would be logical.
This topic has been locked.
Pinned topic Generalizing negation by using groups in rules
Answered question This question has been answered.
Unanswered question This question has not been answered yet.
SystemAdmin 110000D4XK197 Posts
Re: Generalizing negation by using groups in rules2012-12-05T10:36:16ZThis is the accepted answer. This is the accepted answer.You need to create a rule for all these entities. The Group function is different.
I think you are talking about the "Or" operator, which is not supported.