Synonym discovery

Patent metadata

Field	Value
Publication number	US-11714837-B2
Application number	US-202217575159-A
Country	US
Kind code	B2
Filing date	Jan 13, 2022
Priority date	Aug 8, 2011
Publication date	Aug 1, 2023
Grant date	Aug 1, 2023

Abstract

Methods, systems, and computer-readable media are provided for facilitating mapping of semantically similar terms between and among two or more information systems. In particular, to facilitate automatic discovery, establishment, and/or statistical validation of linkages between a plurality of different nomenclatures employed by a plurality of information systems, such as multiple electronic health record systems. In embodiments, the imputation of latent synonymy in corpora comprised of samples of historical records from each system enables automated terminology mapping between disparate systems' records, thereby establishing reliable linkages that may subsequently be utilized for realtime decision support, data mining-based research, or other valuable purposes.

First claim

What is claimed is:

1. One or more non-transitory computer-readable media having computer-usable instructions embodied thereon that, when executed, enable a processor to perform a method of discovering latent relationships in data, said method comprising:
obtaining a first set of records with a first organizational structure and a second set of records with a second organizational structure, wherein at least a portion of the first organizational structure is incompatible with the second organizational structure;
computing at least one non-parameteric matching measure based on a first cluster of raw data comprising a first set of field values stored in a first data field for the first set of records and a second cluster of raw data comprising a second set of field values stored in a second data field for the second set of records;
computing at least one semantic matching measure based on the first cluster of raw data and the second cluster of raw data;
computing a measure of similarity for the first cluster of raw data and the second cluster of raw data based on a weighted combination of the at least one non-parametric matching measure and the at least one semantic matching measure;
determining that the measure of similarity for the first cluster of raw data and the second cluster of raw data is within a threshold measure of similarity; and
provisionally binding the first data field of the first set of records to the second data field of the second set of records in response to the determination that the measure of similarity for the first cluster of raw data stored in the first data field and the second cluster of raw data stored in the second data field is within the threshold measure of similarity,
the provisional binding, of the first data field of the first set of records to the second data field of the second set of records in response to the determination, forming a provisional mapping.

2. The non-transitory computer-readable media of claim 1 , wherein execution of the computer-usable instructions further enables the processor to:
determine an aggregate similarity metric from a vector of weights that forms a dot product with elements of a similarity metric vector.

3. The non-transitory computer-readable media of claim 1 , wherein:
the first set of records includes at least 500 instances of raw data corresponding to the first data field; and
the calculated measure of similarity is based on a two-sample Kolmogorov-Smirnov D test.

4. The non-transitory computer-readable media of claim 1 , wherein execution of the computer-usable instructions further enables the processor to:
change weights associated with the weighted combination based on an amount of instances of raw data that are available.

5. The non-transitory computer-readable media of claim 4 , wherein the calculated measure of similarity is based on a Cramer V test.

6. The non-transitory computer-readable media of claim 1 , wherein execution of the computer-usable instructions further enable the processor to: discard extreme values of the first cluster of raw data.

7. The non-transitory computer-readable media of claim 1 , wherein execution of the computer-usable instructions further enables the processor to: present the provisional mapping for display, thereby permitting a user to modify the provisional mapping by including or excluding terms from the provisional mapping.

8. A method for discovering latent relationships in data, the method comprising:
obtaining a first set of records with a first organizational structure and a second set of records with a second organizational structure, wherein at least a portion of the first organizational structure is incompatible with the second organizational structure;
computing at least one non-parameteric matching measure based on a first cluster of raw data comprising a first set of field values stored in a first data field for the first set of records and a second cluster of raw data comprising a second set of field values stored in a second data field for the second set of records;
computing at least one semantic matching measure based on the first cluster of raw data and the second cluster of raw data;
computing a measure of similarity for the first cluster of raw data and the second cluster of raw data based on a weighted combination of the at least one non-parametric matching measure and the at least one semantic matching measure;
determining that the measure of similarity for the first cluster of raw data and the second cluster of raw data is within a threshold measure of similarity; and
provisionally binding the first data field of the first set of records to the second data field of the second set of records in response to the determination that the measure of similarity for the first cluster of raw data stored in the first data field and the second cluster of raw data stored in the second data field is within the threshold measure of similarity,
the provisional binding, of the first data field of the first set of records to the second data field of the second set of records in response to the determination, forming a provisional mapping.

9. The method of claim 8 , further comprising discarding extreme values of the first cluster of raw data.

10. The method of claim 9 , wherein the first set of records includes at least 500 instances of raw data corresponding to the first data field.

11. The method of claim 8 , further comprising selecting raw data from the second set of records by matching one or more demographic attributes associated with the first set of records to demographic attributes from the second set of records.

12. The method of claim 8 , wherein computing the measure of similarity is based on a two-sample Kolmogorov-Smirnov D test.

13. The method of claim 8 , wherein the measure of similarity comprises a non-parametric metric.

14. The method of claim 13 , wherein the non-parametric metric is associated with a Cramer V test.

15. The method of claim 8 , further comprising presenting the provisional mapping for display, thereby permitting a user to modify the provisional mapping by including or excluding terms from the provisional mapping.

16. A system for discovering latent relationships in data comprising:
one or more processors;
one or more non-transitory computer-readable media having computer-usable instructions embodied thereon that, when executed, enable the one or more processors to perform a method comprising:
obtaining a first set of records with a first organizational structure and a second set of records with a second organizational structure, wherein at least a portion of the first organizational structure is incompatible with the second organizational structure;
computing at least one non-parameteric matching measure based on a first cluster of raw data comprising a first set of field values stored in a first data field for the first set of records and a second cluster of raw data comprising a second set of field values stored in a second data field for the second set of records;
computing at least one semantic matching measure based on the first cluster of raw data and the second cluster of raw data;
computing a measure of similarity for the first cluster of raw data and the second cluster of raw data based on a weighted combination of the at least one non-parametric matching measure and the at least one semantic matching measure;
determining that the measure of similarity for the first cluster of raw data and the second cluster of raw data is within a threshold measure of similarity; and
provisionally binding the first data field of the

Assignees

• Cerner Innovation Inc

Inventors

• Mcnair Douglas S
• Kailasam Kanakasabha K
• Murrish John Christopher

Classifications

• G06F16/288 Primary

  Entity relationship models · CPC title

• G16H10/60 Primary

  for patient-specific data, e.g. for electronic patient records · CPC title

• G16H15/00

  ICT specially adapted for medical reports, e.g. generation or transmission thereof · CPC title

• G16H50/70

  for mining of medical data, e.g. analysing previous cases of other patients · CPC title