Granted Patent B1

US 10,185,809 · App. 14/054,272 · Granted Jan 22, 2019

Abstract

Claims (114)

1. A computer-implemented method, comprising:

obtaining, from a first set of database systems over a distributed communication network and by a data integration module of a computer system, a stream of encrypted transaction data that includes one or more data fields for de-identified transaction profiles of multiple transaction sources that are located within a geographic region, wherein the computer system comprises the data integration module and a prediction module that communicates with the data integration module to determine at least one data projection;

processing, by the data integration module, contents of the stream of encrypted transaction data to generate a structured aggregate data record, the processing comprising:

determining, for each de-identified transaction profile from among the de-identified transaction profiles of the multiple transaction sources, a respective mapping that identifies data fields that correspond to the same transaction profile;

generating, for each de-identified transaction profile and based on analysis of the stream of encrypted transaction data, a linking identifier for identifying transaction profile data corresponding to the same transaction profile and for removing at least a subset of duplicate transaction profile data included among the same transaction profiles;

determining, for each de-identified transaction profile, a transaction source identifier based on the contents of the corresponding data fields that are identified by the respective mapping;

generating, for each de-identified transaction profile and using the respective transaction source identifiers determined for each de-identified transaction profile, a respective database record that includes the corresponding data fields that are identified by the respective mapping and the linking identifier; and

generating the structured aggregate data record, the structured aggregate data record including (i) transaction source identifiers determined for each of the de-identified transaction profiles, and (ii) the respective database records generated for each transaction source identifier, and wherein generating the structured aggregate data record includes removing, based on the linking identifier, a subset of data that is duplicated among the respective database records;

determining, by the prediction module and based on the contents of the structured aggregate data record generated using the data integration module, one or more regression factors used to identify future transactions of the transaction sources within the geographic region;

determining, by the prediction module and using the one or more regression factors, a first data projection that identifies one or more data fields for a set of future transactions of each of the transaction sources located within the geographic region;

obtaining, from a second set of database systems over the distributed communication network and by the prediction module, a stream of distributor transaction data that includes one or more reference data fields associated with each of the transaction sources;

comparing, by the prediction module, the one or more data fields for the set of future transactions identified by the first data projection to the one or more reference fields in the stream of distributor transaction data that are associated with each of the transaction sources;

in response to comparing, identifying a sample bias in the first data projection based on information describing the set of future transactions identified by the first data projection that was determined using the one or mote regression factors, wherein the sample bias limits accuracy or reliability of the first data projection,

determining an adjustment factor that: i) accounts for the sample bias in the first data projection based on information in the stream or distributor transaction data, and ii) improves accuracy or reliability in projections when used to determine a subsequent data projection;

adjusting, by the prediction module, the one or more regression factors based on the adjustment factor that accounts for the sample bias in the first data projection;

determining, by the prediction module and using the one or more adjusted regression factors, a second data projection that identifies one or more additional data fields for the set of future transactions of each of the transaction sources located within the geographic region, wherein the one or more additional data fields are not identified within the first data projection, and wherein the adjustment factor that accounts for the sample bias in the first data projection causes the second data projection to have improved accuracy or reliability relative to the first data projection; and

generating, by a reporting module of the computer system that communicates with the prediction module, a report that includes at least the first and second data projections determined by the prediction module, the report being provided for output to a client device.

2. The method of claim 1 , wherein:

determining the one or more regression factors comprises determining different factors for each database system included within the first set of database systems;

determining the first data projection comprises determining, for each database system included within the first set of database systems, a different projection based on the different regression factors for each database system included within the first set of database systems;

comparing the one or more data fields identified by the first data projection to the one or more reference fields associated with each of the transaction sources comprises comparing the different projections determined for the database systems included within the first set of database systems; and

adjusting the one or more regression factors comprises adjusting the one or more regression factors determined for each database system included within the first set of database systems based on the comparison of the different projections determined for the database systems included within the first set of database systems.

3. The method of claim 1 , further comprising:

receiving claims data from one or more database systems associated with health insurance companies;

comparing one or more components of the second data projection to one or more components of the claims data received from database systems associated with health insurance companies; and

validating the second data projection based on the comparison of the one or more components of the second projection to the one or more components of the claims data received from database systems associated with health insurance companies.

4. The method of claim 1 , wherein:

the de-identified transaction profiles each correspond to a patient located within the geographic region; and

the one or more data fields for the de-identified transaction profiles comprise a data field specifying metastatic and oncological stage data for a patient.

5. The method of claim 1 , wherein:

the geographic region comprises a national area including sub-national segments;

comparing the one or more data fields identified by the first data projection to the one or more reference data fields associated with each of the transaction sources comprises:

identifying, for each of the one or more database fields identified by the first data projection, a particular sub-national segment from among the sub-national segments included in the national area;

identifying, for each of the one or more reference data fields, a particular sub-national segment from among the sub-national segments included in the national area; and

comparing, for each of the sub-national segments included in the national area, database fields and reference data fields identified to a particular sub-national segment.

6. The method of claim 1 , wherein the stream of distributor transaction data obtained from the second set of database systems comprises pharmaceutical distribution data for pharmaceutical products associated with the de-identified transaction profiles of the multiple transaction sources.

7. The method of claim 1 , wherein the multiple transaction sources that are located within the geographic region comprises at least one of a pharmacy, a hospital, or a medical clinic.

8. The method of claim 1 , wherein determining the adjustment factor comprises:

determining the adjustment factor based on a current census view in the distributor transaction data that provides more detailed transaction information relative to the contents of the structured aggregate data record that cause the sample bias in the first data projection, wherein the more detailed transaction information describes item transactions across a region that is more expansive than the geographic region.

9. The method of claim 8 , wherein the adjustment factor accounts for the sample bias and improves accuracy or reliability in projections by:

using the current census view in the distributor transaction data to determine the adjusted regression factors and to refine resulting projections about future transactions that are included in the second data projection.

10. A non-transitory computer-readable storage device encoded with a computer program, the program comprising instructions that when executed by one or more computers cause the one or more computers to perform operations comprising:

obtaining, from a first set of database systems over a distributed communication network and by a data integration module of a computer system, a stream of encrypted transaction data that includes one or more data fields for de-identified transaction profiles of multiple transaction sources that are located within a geographic region, wherein the computer system comprises the data integration module and a prediction module that communicates with the data integration module to determine at least one data projection;

processing, by the data integration module, contents of the stream of encrypted transaction data to generate a structured aggregate data record, the processing comprising:

determining, for each de-identified transaction profile from among the de-identified transaction profiles of the multiple transaction sources, a respective mapping that identifies data fields that correspond to the same transaction profile;

generating, for each de-identified transaction profile and based on analysis of the stream of encrypted transaction data, a linking identifier for identifying transaction profile data corresponding to the same transaction profile and for removing at least a subset of duplicate transaction profile data included among the same transaction profiles;

determining, for each de-identified transaction profile, a transaction source identifier based on the contents of the corresponding data fields that are identified by the respective mapping;

generating, for each de-identified transaction profile and using the respective transaction source identifiers determined for each de-identified transaction profile, a respective database record that includes the corresponding data fields that are identified by the respective mapping and the linking identifier; and

generating the structured aggregate data record, the structured aggregate data record including (i) transaction source identifiers determined for each of the de-identified transaction profiles, and (ii) the respective database records generated for each transaction source identifier, and wherein generating the structured aggregate data record includes removing, based on the linking identifier, a subset of data that is duplicated among the respective database records;

determining, by the prediction module and based on the contents of the structured aggregate data record generated using the data integration module, one or more regression factors used to identify future transactions of the transaction sources within the geographic region;

determining, by the prediction module and using the one or more regression factors, a first data projection that identifies one or more data fields for a set of future transactions of each of the transaction sources located within the geographic region;

obtaining, from a second set of database systems over the distributed communication network and by the prediction module, a stream of distributor transaction data that includes one or more reference data fields associated with each of the transaction sources;

comparing, by the prediction module, the one or more data fields for the set of future transactions identified by the first data projection to the one or more reference fields in the stream of distributor transaction data that are associated with each of the transaction sources;

in response to comparing, identifying a sample bias in the first data projection based on information describing the set of future transactions identified by the first data projection that was determined using the one or mote regression factors, wherein the sample bias limits accuracy or reliability of the first data projection,

determining an adjustment factor that: i) accounts for the sample bias in the first data projection based on information in the stream or distributor transaction data, and ii) improves accuracy or reliability in projections when used to determine a subsequent data projection;

adjusting, by the prediction module, the one or more regression factors based on the adjustment factor that accounts for the sample bias in the first data projection;

determining, by the prediction module and using the one or more adjusted regression factors, a second data projection that identifies one or more additional data fields for the set of future transactions of each of the transaction sources located within the geographic region, wherein the one or more additional data fields are not identified within the first data projection, and wherein the adjustment factor that accounts for the sample bias in the first data projection causes the second data projection to have improved accuracy or reliability relative to the first data projection; and

generating, by a reporting module of the computer system that communicates with the prediction module, a report that includes at least the first and second data projections determined by the prediction module, the report being provided for output to a client device.

11. The device of claim 10 , wherein:

determining the one or more regression factors comprises determining different factors for each database system included within the first set of database systems;

determining the first data projection comprises determining, for each database system included within the first set of database systems, a different projection based on the different regression factors for each database system included within the first set of database systems;

comparing the one or more data fields identified by the first data projection to the one or more reference fields associated with each of the transaction sources comprises comparing the different projections determined for the database systems included within the first set of database systems; and

adjusting the one or more regression factors comprises adjusting the one or more regression factors determined for each database system included within the first set of database systems based on the comparison of the different projections determined for the database systems included within the first set of database systems.

12. The device of claim 10 , wherein the operations further comprise:

receiving claims data from one or more database systems associated with health insurance companies;

comparing one or more components of the second data projection to one or more components of the claims data received from database systems associated with health insurance companies; and

validating the second data projection based on the comparison of the one or more components of the second projection to the one or more components of the claims data received from database systems associated with health insurance companies.

13. The device of claim 10 , wherein:

the de-identified transaction profiles each correspond to a patient located within the geographic region; and

the one or more data fields for the de-identified transaction profiles comprise a data field specifying metastatic and oncological stage data for a patient.

14. The device of claim 10 , wherein:

the geographic region comprises a national area including sub-national segments;

comparing the one or more data fields identified by the first data projection to the one or more reference data fields associated with each of the transaction sources comprises:

identifying, for each of the one or more database fields identified by the first data projection, a particular sub-national segment from among the sub-national segments included in the national area;

identifying, for each of the one or more reference data fields, a particular sub-national segment from among the sub-national segments included in the national area; and

comparing, for each of the sub-national segments included in the national area, database fields and reference data fields identified to a particular sub-national segment.

15. The device of claim 10 , wherein the multiple transaction sources that are located within the geographic region comprises at least one of a pharmacy, a hospital, or a medical clinic.

16. A system comprising:

one or more processors; and

one or more storage mediums storing instructions that, when executed by the one or more processors, cause the one or more processors to perform operations comprising:

obtaining, from a first set of database systems over a distributed communication network and by a data integration module of a computer system, a stream of encrypted transaction data that includes one or more data fields for de-identified transaction profiles of multiple transaction sources that are located within a geographic region, wherein the computer system comprises the data integration module and a prediction module that communicates with the data integration module to determine at least one data projection;

processing, by the data integration module, contents of the stream of encrypted transaction data to generate a structured aggregate data record, the processing comprising:

determining, for each de-identified transaction profile from among the de-identified transaction profiles of the multiple transaction sources, a respective mapping that identifies data fields that correspond to the same transaction profile;

generating, for each de-identified transaction profile and based on analysis of the stream of encrypted transaction data, a linking identifier for identifying transaction profile data corresponding to the same transaction profile and for removing at least a subset of duplicate transaction profile data included among the same transaction profiles;

determining, for each de-identified transaction profile, a transaction source identifier based on the contents of the corresponding data fields that are identified by the respective mapping;

generating, for each de-identified transaction profile and using the respective transaction source identifiers determined for each de-identified transaction profile, a respective database record that includes the corresponding data fields that are identified by the respective mapping and the linking identifier; and

generating the structured aggregate data record, the structured aggregate data record including (i) transaction source identifiers determined for each of the de-identified transaction profiles, and (ii) the respective database records generated for each transaction source identifier, and wherein generating the structured aggregate data record includes removing, based on the linking identifier, a subset of data that is duplicated among the respective database records;

determining, by the prediction module and based on the contents of the structured aggregate data record generated using the data integration module, one or more regression factors used to identify future transactions of the transaction sources within the geographic region;

determining, by the prediction module and using the one or more regression factors, a first data projection that identifies one or more data fields for a set of future transactions of each of the transaction sources located within the geographic region;

obtaining, from a second set of database systems over the distributed communication network and by the prediction module, a stream of distributor transaction data that includes one or more reference data fields associated with each of the transaction sources;

comparing, by the prediction module, the one or more data fields for the set of future transactions identified by the first data projection to the one or more reference fields in the stream of distributor transaction data that are associated with each of the transaction sources;

in response to comparing, identifying a sample bias in the first data projection based on information describing the set of future transactions identified by the first data projection that was determined using the one or mote regression factors, wherein the sample bias limits accuracy or reliability of the first data projection,

determining an adjustment factor that: i) accounts for the sample bias in the first data projection based on information in the stream or distributor transaction data, and ii) improves accuracy or reliability in projections when used to determine a subsequent data projection;

adjusting, by the prediction module, the one or more regression factors based on the adjustment factor that accounts for the sample bias in the first data projection;

determining, by the prediction module and using the one or more adjusted regression factors, a second data projection that identifies one or more additional data fields for the set of future transactions of each of the transaction sources located within the geographic region, wherein the one or more additional data fields are not identified within the first data projection, and wherein the adjustment factor that accounts for the sample bias in the first data projection causes the second data projection to have improved accuracy or reliability relative to the first data projection; and

generating, by a reporting module of the computer system that communicates with the prediction module, a report that includes at least the first and second data projections determined by the prediction module, the report being provided for output to a client device.

17. The system of claim 16 , wherein:

determining the one or more regression factors comprises determining different factors for each database system included within the first set of database systems;

determining the first data projection comprises determining, for each database system included within the first set of database systems, a different projection based on the different regression factors for each database system included within the first set of database systems;

comparing the one or more data fields identified by the first data projection to the one or more reference fields associated with each of the transaction sources comprises comparing the different projections determined for the database systems included within the first set of database systems; and

adjusting the one or more regression factors comprises adjusting the one or more regression factors determined for each database system included within the first set of database systems based on the comparison of the different projections determined for the database systems included within the first set of database systems.

18. The system of claim 16 , wherein the operations further comprise:

receiving claims data from one or more database systems associated with health insurance companies;

comparing one or more components of the second data projection to one or more components of the claims data received from database systems associated with health insurance companies; and

validating the second data projection based on the comparison of the one or more components of the second projection to the one or more components of the claims data received from database systems associated with health insurance companies.

19. The system of claim 16 , wherein:

the de-identified transaction profiles each correspond to a patient located within the geographic region; and

the one or more data fields for the de-identified transaction profiles comprise a data field specifying metastatic and oncological stage data for a patient.

20. The system of claim 16 , wherein:

the geographic region comprises a national area including sub-national segments;

comparing the one or more data fields identified by the first data projection to the one or more reference data fields associated with each of the transaction sources comprises:

identifying, for each of the one or more database fields identified by the first data projection, a particular sub-national segment from among the sub-national segments included in the national area;

identifying, for each of the one or more reference data fields, a particular sub-national segment from among the sub-national segments included in the national area; and

comparing, for each of the sub-national segments included in the national area, database fields and reference data fields identified to a particular sub-national segment.

Assignments (11)

Cited By (5)

US 12,347,537 US 12,406,754 US 12,424,304 US 12,541,798 US 12,633,421