Criteria modification to improve analysis

• US 10,083,161 B2
• Filed: 10/15/2015
• Issued: 09/25/2018
• Est. Priority Date: 10/15/2015
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Freed, Andrew R., Lorenzo, Mario J., Nowicki, Jeffrey B., Pierce, Daniel Z., Von Berge, Jerry L.
• Primary Examiner(s): Baderman, Scott
• Assistant Examiner(s): EDWARDS, JASON T
• Application Number:
  US14/884,217
• Publication Number:
  US 20170109333A1
• Time in Patent Office: 1,076 Days
• Field of Search: 715230
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G06F 40/169 Annotation, e.g. comment data or footnotes
• CPC Class Codes: G06Q 40/08 Insurance
• CPC Class Codes: G16H 10/20 for electronic clinical trials or questionnaires
• CPC Class Codes: G16H 10/60 for patient-specific data, e.g. for electronic patient records
• CPC Class Codes: G16H 40/20 for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16H 50/30 for calculating health indices; for individual health risk assessment

Abstract

Systems, methods, and computer program products to perform an operation comprising receiving a policy document specifying a plurality of criteria, identifying, in a segment of unstructured text in the policy document, a criteria delimiter, creating a first portion and a second portion of the segment of unstructured text, wherein the first and second portions are created based on the criteria delimiter, wherein the first and second portions comprise a first criterion and a second criterion, respectively, and responsive to identifying an anaphora in the second portion of the segment of unstructured text that is related to a term in the first portion of the segment of unstructured text, combining the first criterion and the second criterion into a composite criterion.

Claims

1. A method comprising:

receiving an electronic policy document specifying a plurality of criteria;

identifying, in a segment of unstructured text in the policy document, a criteria delimiter;

identifying a first portion and a second portion of the segment of unstructured text, wherein the first and second portions are identified based on the criteria delimiter, wherein the first and second portions comprise a first criterion and a second criterion, respectively, wherein the first criterion comprises text in the policy document specifying;

a first concept, a value for the first concept, and a first operator for the value for the first concept, wherein the second criterion comprises text in the policy document specifying;

a second concept, a value for the second concept, and a second operator for the value for the second concept;

identifying, by a first annotator of a plurality of annotators, an anaphora in the second portion of the segment of unstructured text that relates the second concept to the first concept;

combining, based on the identified anaphora, the first criterion and the second criterion into a composite criterion, wherein the composite criterion comprises the first and second concepts, at least one of the first and second operators, and at least one of the values for the first and second concepts; and

storing an indication that the composite criterion is one of the plurality of criteria for the policy document, wherein combining the first and second criteria reduces an amount of processing resources required to process cases against the plurality of criteria of the policy document relative to processing cases without combining the first and second criteria.

2. The method of claim 1, wherein the indication is stored as an Unstructured Information Management Architecture (UIMA) Common Analysis System (CAS) object, the method further comprising determining that the first and second criteria are related criteria by:

identifying the first concept in the first portion of the segment of unstructured text and the;

second concept in the second portion of the segment of unstructured text;

determining a distance between a first concept node of an ontology and a second concept node of the ontology, wherein the first and second nodes of the ontology are related to the first and second concepts, respectively;

computing a score for the first concept and the second concept, wherein the similarity score is based on the distance between the first concept node and the second concept node in the ontology and reflects a level of similarity between the first and second concepts; and

prior to combining the first criterion and the second criterion into the composite criterion, determining that the similarity score exceeds a similarity threshold.

3. The method of claim 2, wherein the segment of unstructured text comprises a single criterion prior to identifying the first and second portions of the unstructured text, the method further comprising:

upon determining that (i) the score does not exceed the similarity threshold, and (ii) the anaphora does not relate the second concept to the first concept;

refraining from combining the first and second criteria into a composite criteria; and

storing an indication that the first and second criteria are distinct criteria for the policy document.

4. The method of claim 1, wherein the composite criterion further comprises the first and second operators and the values for the first and second concepts.

5. The method of claim 1, further comprising:

receiving a first case comprising a plurality of attributes;

processing the plurality of attributes of the first case against the plurality of criteria of the policy document;

determining that the at least one of the first and second operators of the composite criterion applied to a first attribute value of a first attribute of the plurality of attributes does not satisfy the at least one of the values for the first and second concepts of the composite criterion; and

determining that the first case does not satisfy the criteria of the policy document.

6. The method of claim 1, further comprising:

identifying, by a second annotator of the plurality of annotators, a plurality of concepts including the first and second concepts in the policy document, wherein the anaphora further relates the value for the second concept to the value for the first concept, wherein the plurality of annotators comprise;

(i) natural language processors, (ii) annotators configured to identify delimiters in electronic text, (iii) annotators configured to identify one or more criteria, (iv) annotators configured to identify anaphora, and (v) annotators configured to identify one or more concepts.

7. The method of claim 1, wherein the criteria delimiter comprises one of:

(i) a period, (ii) a semicolon, (iii) a comma, (iv) a line break, (v) a character, and (vi) a list prefix delimiter.

8. A system, comprising:

a processor; and

a memory containing a program which when executed by the processor performs an operation comprising;

receiving an electronic policy document specifying a plurality of criteria;

identifying, in a segment of unstructured text in the policy document, a criteria delimiter;

identifying a first portion and a second portion of the segment of unstructured text, wherein the first and second portions are identified based on the criteria delimiter, wherein the first and second portions comprise a first criterion and a second criterion, respectively, wherein the first criterion comprises text in the policy document specifying;

a first concept, a value for the first concept, and a first operator for the value for the first concept, wherein the second criterion comprises text in the policy document specifying;

a second concept, a value for the second concept, and a second operator for the value for the second concept;

identifying, by a first annotator of a plurality of annotators, an anaphora in the second portion of the segment of unstructured text that relates the second concept to the first concept;

combining, based on the identified anaphora, the first criterion and the second criterion into a composite criterion, wherein the composite criterion comprises the first and second concepts, at least one of the first and second operators, and at least one of the values for the first and second concepts; and

storing an indication that the composite criterion is one of the plurality of criteria for the policy document, wherein combining the first and second criteria reduces an amount of processing resources required to process cases against the plurality of criteria of the policy document relative to processing cases without combining the first and second criteria.

9. The system of claim 8, wherein the indication is stored as an Unstructured Information Management Architecture (UIMA) Common Analysis System (CAS) object, the operation further comprising determining that the first and second criteria are related criteria by:

identifying the first concept in the first portion of the segment of unstructured text and the;

second concept in the second portion of the segment of unstructured text;

determining a distance between a first concept node of an ontology and a second concept node of the ontology, wherein the first and second nodes of the ontology are related to the first and second concepts, respectively;

computing a score for the first concept and the second concept, wherein the similarity score is based on the distance between the first concept node and the second concept node in the ontology and reflects a level of similarity between the first and second concepts; and

prior to combining the first criterion and the second criterion into the composite criterion, determining that the similarity score exceeds a similarity threshold.

10. The system of claim 9, wherein the segment of unstructured text comprises a single criterion prior to identifying the first and second portions of the unstructured text, the operation further comprising:

upon determining that (i) the score does not exceed the similarity threshold, and (ii) the anaphora does not relate the second concept to the first concept;

refraining from combining the first and second criteria into a composite criteria; and

storing an indication that the first and second criteria are distinct criteria for the policy document.

11. The system of claim 8, wherein the composite criterion further comprises the first and second operators and the values for the first and second concepts.

12. The system of claim 8, the operation further comprising:

receiving a first case comprising a plurality of attributes;

processing the plurality of attributes of the first case against the plurality of criteria of the policy document;

determining that the at least one of the first and second operators of the composite criterion applied to a first attribute value of a first attribute of the plurality of attributes does not satisfy the at least one of the values for the first and second concepts of the composite criterion; and

determining that the first case does not satisfy the criteria of the policy document.

13. The system of claim 8, the operation further comprising:

identifying, by a second annotator of the plurality of annotators, a plurality of concepts including the first and second concepts in the policy document, wherein the anaphora further relates the value for the second concept to the value for the first concept, wherein the plurality of annotators comprise;

(i) natural language processors, (ii) annotators configured to identify delimiters in electronic text, (iii) annotators configured to identify one or more criteria, (iv) annotators configured to identify anaphora, and (v) annotators configured to identify one or more concepts.

14. The system of claim 8, wherein the criteria delimiter comprises one of:

(i) a period, (ii) a semicolon, (iii) a comma, (iv) a line break, (v) a character, and (vi) a list prefix delimiter.

15. A computer program product, comprising:

a computer-readable storage medium having computer-readable program code embodied therewith, the computer-readable program code executable by one or more computer processors to perform an operation comprising;

receiving an electronic policy document specifying a plurality of criteria;

identifying, in a segment of unstructured text in the policy document, a criteria delimiter;

identifying a first portion and a second portion of the segment of unstructured text, wherein the first and second portions are identified based on the criteria delimiter, wherein the first and second portions comprise a first criterion and a second criterion, respectively, wherein the first criterion comprises text in the policy document specifying;

a first concept, a value for the first concept, and a first operator for the value for the first concept, wherein the second criterion comprises text in the policy document specifying;

a second concept, a value for the second concept, and a second operator for the value for the second concept;

identifying, by a first annotator of a plurality of annotators, an anaphora in the second portion of the segment of unstructured text that relates the second concept to the first concept;

combining, based on the identified anaphora, the first criterion and the second criterion into a composite criterion, wherein the composite criterion comprises the first and second concepts, at least one of the first and second operators, and at least one of the values for the first and second concepts; and

storing an indication that the composite criterion is one of the plurality of criteria for the policy document, wherein combining the first and second criteria reduces an amount of processing resources required to process cases against the plurality of criteria of the policy document relative to processing cases without combining the first and second criteria.

16. The computer program product of claim 15, wherein the indication is stored as an Unstructured Information Management Architecture (UIMA) Common Analysis System (CAS) object, the operation further comprising determining that the first and second criteria are related criteria by:

identifying a first concept in the first portion of the segment of unstructured text and a second concept in the second portion of the segment of unstructured text;

determining a distance between a first concept node of an ontology and a second concept node of the ontology, wherein the first and second nodes of the ontology are related to the first and second concepts, respectively;

computing a score for the first concept and the second concept, wherein the similarity score is based on the distance between the first concept node and the second concept node in the ontology and reflects a level of similarity between the first and second concepts; and

prior to combining the first criterion and the second criterion into the composite criterion, determining that the similarity score exceeds a similarity threshold.

17. The computer program product of claim 16, wherein the segment of unstructured text comprises a single criterion prior to identifying the first and second portions of the unstructured text, the operation further comprising:

upon determining that (i) the score does not exceed the similarity threshold, and (ii) the anaphora does not relate the second concept to the first concept;

refraining from combining the first and second criteria into a composite criteria; and

storing an indication that the first and second criteria are distinct criteria for the policy document.

18. The computer program product of claim 15, wherein the composite criterion further comprises the first and second operators and the values for the first and second concepts.

19. The computer program product of claim 15, wherein the criteria delimiter comprises one of:

(i) a period, (ii) a semicolon, (iii) a comma, (iv) a line break, (v) a character, and (vi) a list prefix delimiter, the operation further comprising;

receiving a first case comprising a plurality of attributes;

processing the plurality of attributes of the first case against the plurality of criteria of the policy document;

determining that the at least one of the first and second operators of the composite criterion applied to a first attribute value of a first attribute of the plurality of attributes does not satisfy the at least one of the values for the first and second concepts of the composite criterion; and

determining that the first case does not satisfy the criteria of the policy document.

20. The computer program product of claim 15, the operation further comprising:

identifying, by a second annotator of the plurality of annotators, a plurality of concepts including the first and second concepts in the policy document, wherein the anaphora further relates the value for the second concept to the value for the first concept, wherein the plurality of annotators comprise;

(i) natural language processors, (ii) annotators configured to identify delimiters in electronic text, (iii) annotators configured to identify one or more criteria, (iv) annotators configured to identify anaphora, and (v) annotators configured to identify one or more concepts.