Controlling false match rates in record linkage using extreme value theory.
Murat SariyarAndreas BorgKlaus PommereningPublished in: J. Biomed. Informatics (2011)
Keyphrases
- record linkage
- extreme value theory
- candidate matches
- privacy preserving
- data cleaning
- duplicate detection
- entity resolution
- approximate matching
- linked data
- multiple databases
- database
- census data
- group membership
- case study
- structured data
- relational databases
- preprocessing
- e learning
- machine learning
- data mining
- databases