Method and system for focused multi-blocking to increase link identification rates in record comparison

• US 9,760,654 B2
• Filed: 04/26/2013
• Issued: 09/12/2017
• Est. Priority Date: 04/26/2013
• Status: Active Grant

• Current Assignee: Walmart Apollo, LLC (WalMart Inc.)
• Original Assignee: Wal-Mart Stores Texas LLC (WalMart Inc.)
• Inventors: Ray, Andrew Benjamin, Troutman, Nathaniel Philip
• Primary Examiner(s): Morrison, Jay
• Assistant Examiner(s): Hoang, Ken
• Application Number:
  US13/871,885
• Publication Number:
  US 20140324906A1
• Time in Patent Office: 1,600 Days
• Field of Search: 707737, 707E17014, 707692, 707E17005, 707E17046, 707741, 707722, 707769, 707791, 707792, 707999005, 707E17001, 707E17002, 707E17044, 707E17045, 707E17048, 707E17051, 707E17054, 707E17055, 707E17083, 707690, 707695, 707718, 707747, 707752, 707780, 707796, 707797, 707809, 707822, 707827, 707999002, 707999003, 707999102, 709224, 709215, 709226, 706 12, 706 46, 706 48, 706 52
• CPC Class Codes: G06F 16/9024 Graphs; Linked lists G06F16/9027 takes precedence

Abstract

Techniques for comparing customer records to identify linked customer records pertaining to a single customer entity are provided. The techniques include identifying a target group of electronic customer records having data fields containing data pertaining to a customer, identifying one or more focused blockers identifying a data value for an electronic customer record data field, and analyzing the target group of electronic customer records to identify a focused group of electronic customer records containing the focused blocker data value. The techniques also include comparing pairs of electronic customer records from the focused group of electronic customer records to identify linked records which pertain to a single customer entity.

Claims

1. A computer-implemented method comprising:

a computer system identifying a target group of electronic customer records, each electronic customer record having data fields containing data pertaining to a customer;

the computer system, at one or more master nodes, using a mapping system to divide linking tasks into sub-tasks that are distributed to worker nodes in a multi-level tree structure for parallel processing;

the computer system, at one or more first worker nodes of the worker nodes, identifying a first focused blocker, the first focused blocker identifying a data value for an electronic customer record data field;

the computer system analyzing one or more electronic customer records from within the target group of electronic customer records to, for each electronic customer record within the target group of electronic customer records, identify one or more first focused blocker keys, the one or more first focused blocker keys comprising one or more data values from the each electronic customer record of the target group of electronic customer records corresponding to the data value for the electronic customer record data field;

the computer system further analyzing the one or more electronic customer records from within the target group of electronic customer records, and producing one or more additional focused blocker keys based from the further analysis;

the computer system associating the one or more additional focused blocker keys with the one or more electronic customer records from within the target group of electronic customer records;

the computer system, at the one or more first worker nodes of the worker nodes, analyzing the target group of electronic customer records to identify a first focused group of electronic customer records, the first focused group of electronic customer records comprising;

electronic customer records comprising a first focused blocker data value; and

electronic customer records associated with at least one of the one or more additional focused blocker keys;

the computer system, at the one or more first worker nodes of the worker nodes, comparing pairs of electronic customer records from the first focused group of electronic customer records to identify linked records, each linked record comprising two or more electronic customer records which pertain to a single customer entity;

in parallel to the computer system identifying the first focused blocker, the computer system, at one or more second worker nodes of the worker nodes, identifying a second focused blocker, the second focused blocker identifying a third data value for the electronic customer record data field, wherein the third data value is different than the data value identified by the first focused blocker for the electronic customer record data field;

in parallel to the computer system analyzing the target group of electronic customer records to identify the first focused group of electronic customer records, the computer system, at the one or more second worker nodes of the worker nodes, analyzing the target group of electronic customer records to identify a second focused group of electronic customer records, the second focused group of electronic customer records comprising electronic customer records comprising the third data value; and

in parallel to the computer system comparing the pairs of electronic customer records from the first focused group of electronic customer records, the computer system, at the one or more second worker nodes of the worker nodes, comparing pairs of electronic customer records from the second focused group of electronic customer records to identify second linked records, each linked record of the second linked records comprising two or more electronic customer records which pertain to a second single customer entity; and

the computer system, at the one or more master nodes, using a reduction system to receive and combine responses from the worker nodes for output.

2. The method of claim 1, wherein the first focused blocker value comprises a first data value for a first electronic customer record data field and a second data value for a second electronic customer record data field.

3. The method of claim 2, wherein the first focused group of electronic customer records further comprises electronic customer records comprising the first data value and the second data value, the data value comprising the first and second data values.

4. The method of claim 2, wherein the first focused blocker identifies the first data value and the second data value selected from a group consisting of:

a double metaphone of a last name and a zip code;

the last name and the zip code;

the last name and a house number; and

the house number and the zip code.

5. The method of claim 1, wherein the target group of electronic customer records comprises a first group of electronic customer records obtained from a first source and a second group of electronic customer records obtained from a second source.

6. The method of claim 1, wherein the single customer entity is selected from a group consisting of:

a single customer;

a household; and

a business.

7. The method of claim 1, wherein the method further comprises the computer system comparing all possible pairs of electronic customer records from the first focused group of electronic customer records to identify the linked records.

8. The method of claim 1, wherein:

the first focused blocker data value comprises a first data value for a first electronic customer record data field and a second data value for a second electronic customer record data field;

the first focused group of electronic customer records further comprises electronic customer records comprising the first data value and the second data value;

the first focused blocker identifies the first data value and the second data value selected from a group consisting of;

a double metaphone of a last name and a zip code;

the last name and the zip code;

the last name and a house number; and

the house number and the zip code;

the target group of electronic customer records comprises a first group of electronic customer records obtained from a first source and a second group of electronic customer records obtained from a second source; and

the single customer entity is selected from a group consisting of;

a single customer;

a household; and

a business.

9. The method of claim 1, further comprising:

the computer system collecting information based at least in part on the linked records; and

the computer system modifying one or more of the one or more electronic customer records based at least in part on the information.

10. A computer-implemented method comprising:

a computer system identifying a target group of electronic records, each electronic record having data fields containing data pertaining to one or more events;

the computer system, at one or more master nodes, using a mapping system to divide linking tasks into sub-tasks that are distributed to worker nodes in a multi-level tree structure for parallel processing;

the computer system, at one or more first worker nodes of the worker nodes, identifying a first focused blocker, the first focused blocker identifying a data value for an electronic record data field;

the computer system analyzing one or more electronic records from within the target group of electronic records to, for each electronic customer record within the target group of electronic records, identify one or more first focused blocker keys, the one or more first focused blocker keys comprising one or more data values from the each electronic customer record of the target group of electronic records corresponding to the data value for the electronic record data field;

the computer system further analyzing the one or more electronic records from within the target group of electronic records, and producing one or more additional focused blocker keys based from the further analysis;

the computer system associating the one or more additional focused blocker keys with the one or more electronic records from within the target group of electronic records;

the computer system, at the one or more first worker nodes of the worker nodes, analyzing the target group of electronic records to identify a first focused group of electronic records, the first focused group of electronic records comprising;

electronic records comprising a first focused blocker data value; and

electronic records associated with at least one of the one or more additional focused blocker keys;

the computer system, at the one or more first worker nodes of the worker nodes, comparing pairs of electronic records from the first focused group of electronic records to identify linked records, each linked record comprising two or more electronic records which pertain to a single event of the one or more events;

in parallel to the computer system identifying the first focused blocker, the computer system, at one or more second worker nodes of the worker nodes, identifying a second focused blocker, the second focused blocker identifying a third data value for the electronic record data field, wherein the third data value is different than the data value identified by the first focused blocker for the electronic record data field;

in parallel to the computer system analyzing the target group of electronic records to identify the first focused group of electronic records, the computer system, at the one or more second worker nodes of the worker nodes, analyzing the target group of electronic records to identify a second focused group of electronic customer records, the second focused group of electronic customer records comprising electronic customer records comprising the third data value;

in parallel to the computer system comparing the pairs of electronic records from the first focused group of electronic records, the computer system, at the one or more second worker nodes of the worker nodes, comparing pairs of electronic records from the second focused group of electronic customer records to identify second linked records, each linked record of the second linked records comprising two or more electronic customer records which pertain to a second single customer entity; and

the computer system, at the one or more master nodes, using a reduction system to receive and combine responses from the worker nodes for output.

11. The computer-implemented method of claim 10, wherein:

the first focused blocker data value comprises a first data value for a first electronic record data field and a second data value for a second electronic record data field;

the first focused group of electronic records further comprises electronic records comprising the first data value and the second data value; and

the target group of electronic records comprises a first group of electronic records obtained from a first source and a second group of electronic records obtained from a second source.

12. The computer-implemented method of claim 10, further comprising:

collecting information based at least in part on the linked records; and

modifying one or more of the one or more electronic records based at least in part on the information.

13. A computer system comprising:

a computer system comprising a hardware processor, the computer system programmed to;

identify a target group of electronic customer records, each electronic customer record having data fields containing data pertaining to a customer;

use a mapping system, at one or more master nodes, to divide linking tasks into sub-tasks that are distributed to worker nodes in a multi-level tree structure for parallel processing;

identify, at one or more first worker nodes of the worker nodes, a first focused blocker, the first focused blocker identifying a data value for an electronic customer record data field;

analyze one or more electronic customer records from within the target group of electronic customer records to, for each electronic customer record within the target group of electronic customer records, identify one or more first focused blocker keys, the one or more first focused blocker keys comprising one or more data values from the each electronic customer record of the target group of electronic customer records corresponding to the data value for the electronic customer record data field;

further analyze the one or more electronic customer records from within the target group of electronic customer records, and producing one or more additional focused blocker keys based from the further analysis;

associate the one or more additional focused blocker keys with the one or more electronic customer records from within the target group of electronic customer records;

analyze, at the one or more first worker nodes of the worker nodes, the target group of electronic customer records to identify a first focused group of electronic customer records, the first focused group of electronic customer records comprising;

electronic customer records comprising a first focused blocker data value; and

electronic customer records associated with at least one of the one or more additional focused blocker keys;

compare, at the one or more first worker nodes of the worker nodes, pairs of electronic customer records from the first focused group of electronic customer records to identify linked records, each linked record comprising two or more electronic customer records which pertain to a single customer entity;

in parallel to identifying the first focused blocker, identify, at one or more second worker nodes of the worker nodes, a second focused blocker, the second focused blocker identifying a third data value for the electronic customer record data field, wherein the third data value is different than the data value identified by the first focused blocker for the electronic customer record data field;

in parallel to analyzing the target group of electronic customer records to identify the first focused group of electronic customer records, analyze, at the one or more second worker nodes of the worker nodes, the target group of electronic customer records to identify a second focused group of electronic customer records, the second focused group of electronic customer records comprising electronic customer records comprising the third data value;

in parallel to comparing the pairs of electronic customer records from the first focused group of electronic customer records, compare, at the one or more second worker nodes of the worker nodes, pairs of electronic customer records from the second focused group of electronic customer records to identify second linked records, each linked record of the second linked records comprising two or more electronic customer records which pertain to a second single customer entity; and

using a reduction system, at the one or more master nodes, to receive and combine responses from the worker nodes for output.

14. The system of claim 13, wherein the first focused blocker data value comprises a first data value for a first electronic customer record data field and a second data value for a second electronic customer record data field.

15. The system of claim 14, wherein the first focused group of electronic customer records further comprises electronic customer records comprising the first data value and the second data value, the data value comprising the first and second data values.

16. The system of claim 13, wherein the target group of electronic customer records comprises a first group of electronic customer records obtained from a first source and a second group of electronic customer records obtained from a second source.

17. The system of claim 13, wherein the single customer entity is selected from a group consisting of:

a single customer;

a household; and

a business.

18. The system of claim 14, wherein the first focused blocker identifies the first data value and the second data value selected from a group consisting of:

a double metaphone of a last name and a zip code;

the last name and the zip code;

the last name and a house number; and

the house number and the zip code.

19. The system of claim 13, wherein the computer system is further programmed to compare all possible pairs of electronic customer records from the first focused group of electronic customer records to identify the linked records.

20. The system of claim 13, wherein:

the first focused blocker data value comprises a first data value for a first electronic customer record data field and a second data value for a second electronic customer record data field;

the first focused group of electronic customer records further comprises electronic customer records comprising the first data value and the second data value;

the first focused blocker identifies the first data value and the second data value selected from a group consisting of;

a double metaphone of a last name and a zip code;

the last name and the zip code;

the last name and a house number; and

the house number and the zip code;

the target group of electronic customer records comprises a first group of electronic customer records obtained from a first source and a second group of electronic customer records obtained from a second source; and

the single customer entity is selected from a group consisting of;

a single customer;

a household; and

a business.

21. The system of claim 13, wherein:

the computer system is further programmed to;

collect information based at least in part on the linked records; and

modifying one or more of the one or more electronic customer records based at least in part on the information.