Technical details

The CHeReL currently uses Choicemaker software for linkage. Choicemaker provides for standardisation and parsing, and differs from the classical probabilistic approaches and software primarily in the use of an automated blocking algorithm and machine learning technique for 'scoring' or assigning weights. The system also allows users to make use of stacked data (where multiple values are present in a single field) and includes a "transitivity engine" that allows for a user-specified action in the case of transitive linkage problems - for example, if record A is a high probability match to both B and C, but B and C are low probability matches to each other. More detail on these features is described in Goldberg and Borthwick (2004). Additional advantages of Choicemaker for the CHeReL are that the system is multi-user, matches in one rather than several steps, is capable of matching very large files and can be run on variety of platforms.