Negation scope analysis for negation detection
View Patent ↗Negation scope analysis for negation detection is provided. In various embodiments, a phrase is read from a report collection. The phrase is searched for at least one of a predetermined set of negation keywords. A dependency parse tree is generated of the phrase. The dependency parse tree is traversed starting with the at least one of the predetermined set of negation keywords. Based on the traversal, a plurality of words of the phrase are determined that are spanned by the at least one of the predetermined set of negation keywords.
1. A method comprising:
reading a phrase from a report collection;
searching within the phrase for at least one negation keyword of a predetermined set of negation keywords;
generating a dependency parse tree of the phrase, the dependency parse tree comprising the at least one negation keyword connected to a dependency chain comprising a plurality of words depending from the at least one negation keyword, wherein each of the plurality of words within the dependency chain comprises a directional dependency to another one of the plurality of words;
starting with the at least one negation keyword, traversing the dependency chain by iteratively identifying words within a scope of negation and ceasing traversing when the scope of negation becomes stable; and
based on the traversal of the dependency chain, determining a noun phrase within the plurality of words spanned by the at least one negation keyword.
2. The method of claim 1 , wherein traversing the dependency chain comprises adding to the predetermined set of negation keywords at least one additional negation keyword.
3. The method of claim 1 , wherein searching within the phrase comprises applying one or more regular expressions to the phrase.
4. The method of claim 1 , wherein the predetermined set of negation keywords is generated by vocabulary learning.
5. The method of claim 1 , wherein the predetermined set of negation keywords comprises at least one term and its antonym.
6. A computer program product for negation detection, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform a method comprising:
reading a phrase from a medical report collection;
searching within the phrase for at least one negation keyword of a predetermined set of negation keywords;
generating a dependency parse tree of the phrase, the dependency parse tree comprising the at least one negation keyword connected to a dependency chain comprising a plurality of words depending from the at least one negation keyword, wherein each of the plurality of words within the dependency chain comprises a directional dependency to another one of the plurality of words;
starting with the at least one negation keyword, traversing the dependency chain by iteratively identifying words within a scope of negation and ceasing traversing when the scope of negation becomes stable; and
based on the traversal of the dependency chain, determining a noun phrase within the plurality of words spanned by the at least one negation keyword.
7. The computer program product of claim 6 , wherein traversing the dependency chain comprises adding to the predetermined set of negation keywords at least one additional negation keyword.
8. The computer program product of claim 6 , wherein searching within the phrase comprises applying one or more regular expressions to the phrase.
9. The computer program product of claim 6 , wherein the predetermined set of negation keywords is generated by vocabulary learning.
10. The computer program product of claim 6 , wherein the predetermined set of negation keywords comprises at least one term and its antonym.
11. A system comprising:
a data store comprising a medical report collection;
a computing node comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor of the computing node to cause the processor to perform a method comprising:
reading a phrase from the medical report collection;
searching within the phrase for at least one negation keyword of a predetermined set of negation keywords;
generating a dependency parse tree of the phrase, the dependency parse tree comprising the at least one negation keyword connected to a dependency chain extending from the at least one negation keyword, wherein each of the plurality of words within the dependency chain comprises a directional dependency to another one of the plurality of words;
starting with the at least one negation keyword, traversing the dependency chain by iteratively identifying words within a scope of negation and ceasing traversing when the scope of negation becomes stable;
based on the traversal of the dependency chain, determining a noun phrase within the plurality of words spanned by the at least one negation keyword.
12. The system of claim 11 , wherein traversing the dependency chain comprises adding to the predetermined set of negation keywords at least one additional negation keyword.
13. The system of claim 11 , wherein searching within the phrase comprises applying one or more regular expressions to the phrase.
14. The system of claim 11 , wherein the predetermined set of negation keywords is generated by vocabulary learning.