|
Activity Number:
|
61
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 3, 2008 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Section on Statistical Computing
|
| Abstract - #300470 |
|
Title:
|
Probabilistic and Fuzzy Matching as Applied to Record Linkage and Computerized Coding
|
|
Author(s):
|
Michael J. Wenzowski*+
|
|
Companies:
|
Statistics Canada
|
|
Address:
|
R.H. Coats Building, Section 14-A, Ottawa, ON, K1A 0T6, Canada
|
|
Keywords:
|
Record Linkage ; Probabilistic Linkage ; Fuzzy Matching ; Automated Coding ; Computerized Coding ; Coding
|
|
Abstract:
|
As the overall speed of computer processing increases, the general perception of record linkage as an unwieldy and time-consuming process diminishes. Similarly, complex coding tasks which have traditionally been viewed as bordering on problematic are now seeing their implementation possibilities evolve towards the plausible. Each of these problem domains employ similar approaches to identifying the "best" candidate for either linkage or code assignment. This paper describes Statistics Canada's recent experiences in redeveloping their own generalized record linkage and computerized coding software, and highlights the convergence and associated overlap in techniques which has occurred across these two statistical processing domains. The overall system architecture employed, the techniques and algorithms implemented, and examples of the user interface developed are also presented.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2008 program |
