List manipulation in natural language processing

1. A computer-implemented method comprising:

receiving an input text comprising a plurality of lists associated with at least one clinical trial and including a first list indicating various stages of cancer, wherein the first list comprises a first list item, a first comma, a second list item, a second comma, and a third list item;

segmenting the input text into tokens;

annotating the tokens for conjunctions;

annotating the tokens for prefix modifiers and suffix modifiers;

modifying the plurality of lists to generate modified input text, wherein the modified input text expands the first list to explicitly state each respective list item by:

appending a suffix modifier to the first list item;

appending a prefix modifier and the suffix modifier to the second list item; and

appending the prefix modifier to the third list item;

storing the modified input text in a database of clinical trials, wherein the modified input text includes the first list item appended with the suffix modifier, the second list item appended with the prefix modifier and the suffix modifier, and the third list item appended with the prefix modifier;

matching the modified input text to one or more patient profiles in a database of patient profiles by matching stages of cancer associated with the patient profiles to one selected from a group consisting of: the first list item appended with the suffix modifier, the second list item appended with the prefix modifier and the suffix modifier, and the third list item appended with the prefix modifier;

scoring respective matches; and

outputting scored patient profiles to a user interface, wherein patient profiles having a score above a threshold are identified as eligible for the at least one clinical trial and patient profiles having a score below the threshold are identified as not eligible for the at least one clinical trial.

2. The method of claim 1 , wherein annotating tokens for parts of speech further comprises:

identifying respective lists of the plurality of lists based on respective tokens annotated as conjunctions;

identifying respective list items based on respective tokens annotated as conjunctions by:

identifying a first discrete list item based on the first discrete list item being located adjacent to a first token annotated as a comma; and

identifying a first implied list item based on a second token annotated as a dash located between a first number and a second number.

3. The method of claim 1 , wherein a conjunction is selected from the group consisting of: a respective word from a set of conjunction words, a comma, and a dash.

4. The method of claim 1 , further comprising:

annotating a lemma form to respective portions of the input text containing any one of a subset of terms related to the lemma form, wherein the lemma form is associated with a trigger, wherein annotating the lemma form further comprises:

identifying a scope associated with the trigger, wherein the scope comprises a term associated with any one of the subset of terms related to the lemma form, wherein the scope limits a meaning of the trigger.

5. The method of claim 1 , further comprising:

annotating at least a portion of the input text for domain specific content.

6. The method of claim 1 , wherein modifying the plurality of lists further comprises:

matching a portion of the input text to an implied list indicator, wherein the implied list indicator is associated with a first explicit list;

replacing, in a second list of the plurality of lists, the implied list indicator with the first explicit list by:

appending the first explicit list to the input text at a location containing the implied list indicator, wherein the first explicit list comprises a plurality of list items; and

deleting the implied list indicator from the input text.

7. The method of claim 1 , wherein modifying the plurality of lists further comprises:

identifying extraneous text of the input text; and

deleting extraneous text from the input text.

8. A system comprising:

a processor; and

a computer-readable storage medium storing program instructions which, when executed by the processor, cause the processor to perform a method comprising:

receiving an input text comprising a plurality of lists associated with at least one clinical trial and including a first list indicating various stages of cancer, wherein the first list comprises a first list item, a first comma, a second list item, a second comma, and a third list item;

segmenting the input text into tokens;

annotating the tokens for conjunctions;

annotating the tokens for prefix modifiers and suffix modifiers;

modifying the plurality of lists to generate modified input text, wherein the modified input text expands the first list to explicitly state each respective list item by:

appending a suffix modifier to the first list item;

appending a prefix modifier and the suffix modifier to the second list item; and

appending the prefix modifier to the third list item;

storing the modified input text in a database of clinical trials, wherein the modified input text includes the first list item appended with the suffix modifier, the second list item appended with the prefix modifier and the suffix modifier, and the third list item appended with the prefix modifier;

matching the modified input text to one or more patient profiles in a database of patient profiles by matching stages of cancer associated with the patient profiles to one selected from a group consisting of: the first list item appended with the suffix modifier, the second list item appended with the prefix modifier and the suffix modifier, and the third list item appended with the prefix modifier;

scoring respective matches; and

outputting scored patient profiles to a user interface, wherein patient profiles having a score above a threshold are identified as eligible for the at least one clinical trial and patient profiles having a score below the threshold are identified as not eligible for the at least one clinical trial.

9. The system of claim 8 , wherein annotating tokens for parts of speech further comprises:

identifying respective lists of the plurality of lists based on respective tokens annotated as conjunctions;

identifying respective list items based on respective tokens annotated as conjunctions by:

identifying a first discrete list item based on the first discrete list item being located adjacent to a first token annotated as a comma; and

identifying a first implied list item based on a second token annotated as a dash located between a first number and a second number.

10. The system of claim 8 , wherein a conjunction is selected from the group consisting of: a respective word from a set of conjunction words, a comma, and a dash.

11. The system of claim 8 , the processor configured to perform a method further comprising:

annotating a lemma form to respective portions of the input text containing any one of a subset of terms related to the lemma form, wherein the lemma form is associated with a trigger, wherein annotating the lemma form further comprises:

identifying a scope associated with the trigger, wherein the scope comprises a term associated with any one of the subset of terms related to the lemma form, wherein the scope limits a meaning of the trigger.

12. The system of claim 8 , the processor configured to perform a method further comprising:

annotating at least a portion of the input text for domain specific content.

13. The system of claim 8 , wherein modifying the plurality of lists further comprises:

matching a portion of the input text to an implied list indicator, wherein the implied list indicator is associated with a first explicit list;

replacing, in a second list of the plurality of lists, the implied list indicator with the first explicit list by:

appending the first explicit list to the input text at a location containing the implied list indicator, wherein the first explicit list comprises a plurality of list items; and

deleting the implied list indicator from the input text.

14. The system of claim 8 , wherein modifying the plurality of lists further comprises:

identifying extraneous text of the input text; and

deleting extraneous text from the input text.

15. A computer program product comprising a computer readable storage medium having program instructions embodied therewith, wherein the computer readable storage medium is not a transitory signal per se, the program instructions executable by a processor to cause the processor to perform a method comprising:

receiving an input text comprising a plurality of lists associated with at least one clinical trial and including a first list indicating various stages of cancer, wherein the first list comprises a first list item, a first comma, a second list item, a second comma, and a third list item;

segmenting the input text into tokens;

annotating the tokens for conjunctions;

annotating the tokens for prefix modifiers and suffix modifiers;

modifying the plurality of lists to generate modified input text, wherein the modified input text expands the first list to explicitly state each respective list item by:

appending a suffix modifier to the first list item;

appending a prefix modifier and the suffix modifier to the second list item; and

appending the prefix modifier to the third list item;

storing the modified input text in a database of clinical trials, wherein the modified input text includes the first list item appended with the suffix modifier, the second list item appended with the prefix modifier and the suffix modifier, and the third list item appended with the prefix modifier;

matching the modified input text to one or more patient profiles in a database of patient profiles by matching stages of cancer associated with the patient profiles to one selected from a group consisting of: the first list item appended with the suffix modifier, the second list item appended with the prefix modifier and the suffix modifier, and the third list item appended with the prefix modifier;

scoring respective matches; and

outputting scored patient profiles to a user interface, wherein patient profiles having a score above a threshold are identified as eligible for the at least one clinical trial and patient profiles having a score below the threshold are identified as not eligible for the at least one clinical trial.

16. The computer program product of claim 15 , wherein annotating tokens for parts of speech further comprises:

identifying respective lists of the plurality of lists based on respective tokens annotated as conjunctions;

identifying respective list items based on respective tokens annotated as conjunctions by:

identifying a first discrete list item based on the first discrete list item being located adjacent to a first token annotated as a comma; and

identifying a first implied list item based on a second token annotated as a dash located between a first number and a second number.

17. The computer program product of claim 15 , wherein a conjunction is selected from the group consisting of: a respective word from a set of conjunction words, a comma, and a dash.

18. The computer program product of claim 15 , the processor configured to perform a method further comprising:

annotating a lemma form to respective portions of the input text containing any one of a subset of terms related to the lemma form, wherein the lemma form is associated with a trigger, wherein annotating the lemma form further comprises:

identifying a scope associated with the trigger, wherein the scope comprises a term associated with any one of the subset of terms related to the lemma form, wherein the scope limits a meaning of the trigger.

19. The computer program product of claim 15 , the processor configured to perform a method further comprising:

annotating at least a portion of the input text for domain specific content.

20. The computer program product of claim 15 , wherein modifying the plurality of lists further comprises:

matching a portion of the input text to an implied list indicator, wherein the implied list indicator is associated with a first explicit list;

replacing, in a second list of the plurality of lists, the implied list indicator with the first explicit list by:

appending the first explicit list to the input text at a location containing the implied list indicator, wherein the first explicit list comprises a plurality of list items; and

deleting the implied list indicator from the input text.