Bagging, bumping, multiview, and active learning for record linkage with empirical results on patient identity data.
Murat SariyarAndreas BorgPublished in: Comput. Methods Programs Biomed. (2012)
Keyphrases
- record linkage
- active learning
- data sets
- database
- data quality
- data analysis
- information retrieval
- data processing
- multiple databases
- original data
- data sources
- training data
- data points
- knowledge discovery
- database systems
- privacy preserving
- raw data
- metadata
- information systems
- learning algorithm
- data cleaning
- machine learning