Record linkage and the matching process
Record linkage is the methodology of identifying records that correspond to the same entity such as a person, household, or product.
In practice, you compare record pairs and classify them into one of these sets: matched pairs and nonmatched pairs.
Statistical methods of record linkage are required, due to these
reasons:
- Columns contain errors or missing values.
- Data can be unreliable.
- You want to find the matches with a reasonable statistical assurance.