I1 (2.5 points)
https://knobit.org/id/task/rmd-matching/I1
Also as Turtle · JSON-LD · RDF/XML
| dcterms:license | https://creativecommons.org/licenses/by/4.0/ |
|---|---|
| owl:sameAs | https://bitbucket.org/reffa/ontology/id/task/rmd-matching/I1 |
| rdfs:label | I1 (2.5 points) en |
| schema:hasPart | rmd-matching I1: a shared family e-mail as a deterministic key |
| schema:hasPart | rmd-matching I1: first-hit rule order |
| schema:hasPart | rmd-matching I1: blocking on surname loses the married name |
| schema:hasPart | rmd-matching I1: no false-merge rate measured |
| schema:hasPart | rmd-matching I1: personal code required of card-system records |
| schema:identifier | I1 |
| schema:position | 3 |
| schema:text | 0.5 each. (a) MR-001 e-mail exact → auto-merge, but families share an e-mail (P-05): not a deterministic key. (b) Rules applied in order with the first hit deciding: all rules scored, then banded. (c) Blocking on surname alone keeps a woman's records apart after her surname changed at marriage (P-06, P-07): extra blocking passes (personal code; date of birth + first name). (d) No false-merge or false-split rate is measured: a labelled sample, before go-live and periodically. (e) MR-002 requires the personal code of card-system records, which have none: scoped to sources that hold it. en |