- Added support for the
onandon_blockingarguments in theblocking()andpair_ann()functions. - Fixed errors in the evaluation functions.
- Updated
est_block_errorfunction.
- Fixed CRAN errors.
- Awesome Official Statistics badge added.
- Removed unnecessary dependency on the
RcppAlgospackage.
- Added support for word embeddings.
- Updated controls for the ANN algorithms.
- Updated evaluation metrics.
- Added 4 data sets.
- Added 2 new vignettes.
- Added estimators for FPR and FNR due to blocking in record linkage, as proposed by Dasylva and Goussanou (2021).
- Updated examples and documentation.
- Supports the following packages:
RcppHNSW,mlpackandRcppAnnoy. - Supports blocking for deduplication and record linkage.
- Metrics when true blocking is known based on
igraph::compare. - Testing with the
tinytestpackage. - Initial support for the
reclin2package. - Class
blockingintroduced. - S3method for printing.
- First vignette added.
- Evaluation with standard metrics (recall, fpr, etc.) added, works with vector for deduplication.
- Added saving index for hnsw and annoy.
rnndescendsupport added.