Table of Contents

Search

  1. Preface
  2. Introduction
  3. The Design Issues
  4. Standard Population Choices
  5. Parsing, Standardization and Cleaning
  6. Customer Identification Systems
  7. Fraud and Intelligence Systems
  8. Marketing Systems
  9. Simple Search
  10. Summary

Application and Database Design Guide

Application and Database Design Guide

What Identification Data do we Match with

What Identification Data do we Match with

In addition to the words and codes in Names, Addresses, Titles and Descriptions, we frequently use other data to make decisions about whether we believe two reports or records are about the same entity.
Search for
ROBERT J.
JOHNSTON
12 RIVER SIDE
SPRINGVALE
(807) 2334 657
1962/02/12
Yes
BOB JAMES
JOHNSTON
SPRING VALE
2334 657
1962
May be
MR. R. J. JOHNSTONE
35 CITYVIEW CT. SPRINGVALE
1 807 4456 721
1962/12/02
No
ROBERT
JOHNSON
2 MAPLE RD.
BROOKFORD
555 763 2413
1973/10/04
Data such as dates of birth, dates of contract, ages, phone numbers and identity numbers are all subject to error and variation.
When a name is used to bring up candidates on a screen, people use all of the identification data returned to choose whether the records displayed are relevant or appropriate. In automated matching systems, the system itself has to be able to use the same data that people would use.
When people make choices about whether things match or not, they compensate for the error and variation. Our systems have to achieve the very same compensation that people make.
To confirm that records are in fact matching requires that our systems use the same data in the same manner as the human users of our systems would use. In fact our systems need to mimic our very best users doing the same job.

0 COMMENTS

We’d like to hear from you!