Mere-parsing with boundary and semantic driven scoping

• US 10,354,005 B2
• Filed: 07/31/2018
• Issued: 07/16/2019
• Est. Priority Date: 04/13/2007
• Status: Active Grant

• Current Assignee: Optum360, LLC (UnitedHealth Group Incorporated)
• Original Assignee: Optum360, LLC (UnitedHealth Group Incorporated)
• Inventors: Heinze, Daniel T., Morsch, Mark L.
• Primary Examiner(s): Huynh, Cong-Lac
• Application Number:
  US16/050,718
• Publication Number:
  US 20180341636A1
• Time in Patent Office: 350 Days
• Field of Search: 715256, 715234
• CPC Class Codes: G06F 40/205 Parsing
• CPC Class Codes: G06F 40/211 Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
• CPC Class Codes: G06F 40/289 Phrasal analysis, e.g. finite state techniques or chunking
• CPC Class Codes: G06F 40/30 Semantic analysis

Abstract

Methods, systems and computer program products for implementing a mere-parser are disclosed. Text data is processed to generate one or more parse items. A boundary based attribute associated with one of the parse items is identified, and the identified mere attribute is associated with one or more of the remaining parse items that is not blocked from associated with the boundary based attribute.

Claims

1. A method for building a tree of parse items, the method comprising the steps of:

generating a merged parse item from at least two parse items, wherein generating comprises;

comparing a first parse item against a predetermined set of target semantic data to obtain a first semantic match;

comparing a second parse item against the predetermined set to obtain a second semantic match;

comparing a combined text of the first and second parse items against the predetermined set to obtain a merged semantic match;

determining whether the merged semantic match is better than either the first or second semantic matches by testing semantic goodness; and

merging the first and second parse items responsive to determining that the merged semantic match is better; and

forming a portion of a tree data structure such that the merged parse item is a parent of the at least two parse items.

2. The method of claim 1, wherein the first and second parse items are associated with boundary based attributes.

3. The method of claim 2, wherein the boundary based attributes include one or more indicators of certainty or temporality.

4. The method of claim 2, further comprising resolving a conflict between the boundary based attributes associated with the first and second parse items by referencing a priority hierarchy for a class of semantic attributes.

5. The method of claim 1, wherein merging the first and second parse items comprises:

creating a new parse item as a copy of the first parse item; and

merging the new parse item with the second parse item.

6. The method of claim 1, wherein forming the portion of the tree data structure further comprises accumulating ordinal modifiers in a list.

7. The method of claim 1, further comprising:

comparing a third parse item against the predetermined set to obtain a third semantic match;

comparing a combined text of the first, second and third parse items against the predetermined set to obtain a merged semantic match;

determining whether the merged semantic match is better than the first, second or third semantic matches by testing semantic goodness; and

merging the first, second and third parse items responsive to determining that the merged semantic match is better.

8. A computer-implemented mere-parser system comprising:

a semantic test application configured to test a semantic goodness of a merged semantic match;

a semantic data storage configured to store semantic data received from the semantic test application; and

a source data analysis unit comprising a mere-parser application configured to normalize source data, wherein the mere-parser application comprises a parse item bounding system that comprises instructions that when executed cause a processor to;

normalize formatting of the source data;

perform morphological processing on the source data;

normalize special phrases of the source data;

identify boundaries and form parse items from the source data; and

identify propagating attributes of the source data.

9. The system of claim 8, wherein the processor is caused to normalize formatting of the source data by converting source input text to a plain text format.

10. The system of claim 8, wherein the processor is caused to perform morphological processing on the source data by morphing words, numbers and acronyms constituting the source data into one or more predetermined standardized formats.

11. The system of claim 8, wherein the special phrases comprise temporal descriptions, spatial descriptions, medication dosages, or combinations thereof.

12. The system of claim 8, wherein the processor is caused to identify propagating attributes of the source data by iterating through a linked list of parse items.

13. The system of claim 12, wherein the processor is further configured to determine a direction of propagation for each parse item within the linked list associated with a propagating attribute.

14. The system of claim 13, further comprising determining whether a next parse item in the direction of propagation blocks an association of an identified propagating attribute.