ene
FEB
mar
4
2009
2010
2011
1 capturas
4 feb 10 - 4 feb 10
Cerrar
Ayuda
Eustat
Home
Identify
Accessibility
Map
Help
Contact
euskara
|
español
| english
Bindings
Services
Organization
Data-Bank
Themes
You are in:
Home
Services
International Estatistical Seminar
Seminar
Seminar
FUSIÓN DE REGISTROS / ERREGISTROEN BATERATZEA / RECORD LINKAGE
CONTENTS
SUMMARY
Introduction
Definition
Terminology
Uses
Capture-Recapture
Record Linkage Basics
Context
Deterministic Record Linkage
Probabilistic Record Linkage
Not Statistical Matching
Need for Automated Record Linkage
Rec. Link. Theory: Fellegi & Sunter
Basic Definitions and Notation
Agreement Patterns
Example Comparison Space
Conditional Probabilities
Linkage Rule
Error Rates
Clerical Region
Fundamental Theorem
Weight Distribution for Matches
Weight Distribution for Non-Matches
Idealized Distributions
Error Rates, Clerical Review Region
Conditional Independence Assumption
Conditional Independence Example
Fellegi-Sunter Summary
Record Linkage Methodology
Choosing Parameters
Informal Methods
EM Algorithm
Likelihood Function
Complete-data Likelihood Function
Expectation Step
Maximization Step
EM Algorithm
EM Algorithm Remarks
Blocking
Blocking Criteria
Record Linkage Refinements
String Comparator
String Comparator Context
Some String Comparator Types
Bigrams
Jaro-Winkler Comparator
Jaro-Winkler Example
Jaro-Winkler Variations
Similar Characters
Common Prefix
Long String Adjustment
Jaro-Winkler Comparator
Edit Distance String Comparators
Edit Distance Algorithm
Edit Distance Similarity Function
Edit Distance Example
Longest Common Subsequence
LCS Similarity Function
Combination Similarity Function
Evaluating String Comparators
Results of String Comparator Evaluation
Jaro-Winkler Anomaly
Hybrid Comparator
String Comparator Summary
More Than Two Latent Classes
EM for Three Classes
More Than Two Comparison Values
One-to-one Matching
Linear Assignment Algorithm
Error Rates
Practical Considerations
False Non-Match Rate
False Match Rate
Bellin-Rubin
Larsen
Improved Parameter Estimates
Extended Likelihood Function
Larsen, Rubin
Winkler
Data Preparation
Basic Preparation
Address Parsing
Business Lists
Example of Business Name Parsing
Two Kinds of Standardizer
Rule-Based Standardizer
Hidden Markov Standardizer
Hidden Markov Standardizer Reference
Hidden Markov Model
Viterbi Algorithm
HMM Diagram
Standardization Summary
U.S. Census Bureau Software
Matching Programs: Matcher
Matching Programs: Bigmatch
Auxiliary Programs: Counter
Auxiliary Programs: EM
Auxiliary Programs, Standardizer
Services Guide
Press service
Maps and street-guide
International Statistical Seminar
Summer courses
List of products
Statistics Dictionary
Information Service
Technical assistance and methodological advice
Microdata file
Training
Complaint and suggestion
Download
Contracts and contents
Director Welcome
Mission to 2012
Basque Statistics System
Legal Dispositions
Basque Statistics Board. Annual Report 2006
Good Practice Code
International and Cooperation Relations
XX Anniversary of Eustat
Thematic statistical Series
About Data Bank
Maps
Population
Economy
Society
Territory and Environment
Information Society,R+D+i