Package: joinery 1.0.1

Eduard Brüll

joinery: Heuristic Index-Based Record Linkage

Links records that refer to the same entity across sources that share no common key, such as people, firms, or addresses with spelling variation, abbreviations, or reordered words. Linkage is described declaratively as a strategy that normalises, tokenises, phonetically encodes, weights, and blocks each field; candidate pairs are then scored by the rarity-weighted overlap of their tokens and every score is attributed back to individual tokens for explainability. Strategies compose into staged pipelines of exact, fuzzy, and optional embedding-based matching that carry unmatched records forward and resolve entities as connected components. The same strategy runs on an in-memory 'data.table' backend or an out-of-core 'DuckDB' backend, and diagnostic and calibration tools help tune a strategy and filter false positives. The token-retrieval heuristic follows Doherr (2023) <doi:10.2139/ssrn.4326848>.

Authors:Eduard Brüll [aut, cre]

joinery_1.0.1.tar.gz
joinery_1.0.1.tar.gz(r-4.7-any)joinery_1.0.1.tar.gz(r-4.6-any)
joinery_1.0.1.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
joinery/json (API)

# Install 'joinery' in R:

install.packages('joinery', repos = c('https://cran.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/edubruell/joinery/issues

Pkgdown/docs site:https://edubruell.github.io

Datasets:

base_example - Base dataset for record linkage example
match_labels_example - Labelled candidate pairs for calibration examples
street_stopwords - Multilingual Street-Name Stopwords
street_types - Multilingual Street-Type Normalization Dictionary
target_example - Target dataset for record linkage example
workshop_listings - Workshop external directory (target) for record linkage examples
workshop_panel - Multi-year workshop panel for cross-year linkage examples
workshop_register - Workshop guild register (base) for record linkage examples

On CRAN:

2.70 score 9 scripts 79 exports 21 dependencies

Last updated from:bb1fe41906. Checks:4 OK. Indexed: yes.

Target	Result	Time
linux-devel-x86_64	OK	260
source / vignettes	OK	233
linux-release-x86_64	OK	235
wasm-release	OK	174

Exports:ambiguity_plot apply_filter approximate_date as_cologne as_metaphone as_soundex audit_strategy batch_map block_on_tokens block_size_plot calibrate calibrate_matches clear_embedding_cache cluster_size_plot compare_stages compute_embeddings compute_rarity contribution_plot coverage_plot date_tokens deduplicate_table detect_duplicates drop_joinery_temp_tables drop_numeric_tokens drop_short_tokens duckdb_batch_plan duckdb_control embedding_strategy exact_strategy explain_match export_for_labelling extract_initials extract_unmatched filter_stopwords find_stopwords fit_filter frontier_plot fuzzy_tokens generate_ngrams import_labels inspect_tokens joinery_recipe match_features materialize_records multi_stage_dedup multi_stage_search norm_plot normalize_date normalize_street normalize_text numeric_tokens plan_strategy prepare_search_data rarity_distribution rarity_histogram recommendations resolve_entities sample_matches score_density score_embeddings score_histogram search_candidates search_strategy similarity_histogram smooth_rip_identity smooth_rip_log smooth_rip_offset smooth_rip_softmax stage_coverage_plot stage_score_plot strip_vowels summarise_matches token_contribution_plot token_frequency_plot token_shapes top_gap_density use_dictionary vocab_overlap_plot word_tokens

Dependencies:BH cli cpp11 data.table generics glue igraph lattice lifecycle lubridate magrittr Matrix phonics pkgconfig Rcpp rlang S7 stringi timechange tinyplot vctrs

Getting started with joinery

Rendered fromjoinery.Rmdusingknitr::rmarkdown

Last update: 2026-07-07
Started: 2026-07-07

Help page	Topics
Bar chart of candidates-per-record distribution (candidates only)	ambiguity_plot
Apply a fitted filter to match features	apply_filter
Approximate dates by rounding to coarser time units	approximate_date
Encode text phonetically with the Cologne procedure	as_cologne
Encode text phonetically with Metaphone	as_metaphone
Encode text phonetically with Soundex	as_soundex
Audit a Search Strategy Against Data	audit_strategy
Base dataset for record linkage example	base_example
Apply a function to DuckDB table batches	batch_map
Block on a Column's Rare Tokens (region-free blocking)	block_on_tokens
Bar chart of block sizes (requires block_by on strategy)	block_size_plot
Evaluate a fitted filter on labelled pairs	calibrate
Calibrate matches end-to-end (features -> filter -> apply)	calibrate_matches
Clear the embedding reuse cache	clear_embedding_cache
Bar chart of cluster-size distribution (duplicates only)	cluster_size_plot
Compare Stages of a Multi-Stage Match	compare_stages
Compute Embeddings for Records	compute_embeddings
Compute Token Rarity for Record Linkage	compute_rarity
Horizontal bar chart of per-column score contributions	contribution_plot
Bar chart of match coverage (base and/or target)	coverage_plot
Extract date components as tokens	date_tokens
Deduplicate a Table	deduplicate_table
Detect Duplicate Records	detect_duplicates
Drop all temporary DuckDB tables created by joinery	drop_joinery_temp_tables
Drop numeric (house-number) tokens from token lists	drop_numeric_tokens
Drop short tokens from token lists	drop_short_tokens
Create a Batch Plan for DuckDB Table Processing	duckdb_batch_plan
DuckDB Execution Control	Duckdb_Control duckdb_control
Create an Embedding Strategy	Embedding_Strategy embedding_strategy
Define an Exact Matching Strategy	Exact_Strategy exact_strategy
Explain a Single Match	explain_match
Export a match sample to CSV for manual labelling	export_for_labelling
Extract initials from tokens	extract_initials
Extract Unmatched Records	extract_unmatched
Filter out stopwords from token lists	filter_stopwords
Discover candidate stopwords from a prepared token table	find_stopwords
Fit a false-positive filter on labelled match pairs	fit_filter
Cost/recall frontier scatter for a strategy plan	frontier_plot
Collapse near-duplicate tokens to a canonical form	fuzzy_tokens
Generate character n-grams from text	generate_ngrams
Import a labelled CSV back into a feature/label table	import_labels
Inspect Tokens for a Specific Column	inspect_tokens
Build a tidymodels recipe for calibration features	joinery_recipe
Build a per-pair feature table for calibration	Match_Features match_features
Labelled candidate pairs for calibration examples	match_labels_example
Materialize Records by ID	materialize_records
Staged Duplicate Detection (within one table)	multi_stage_dedup
Staged Search Across Tables or Sources	multi_stage_search
Bar chart of embedding norm quantiles	norm_plot
Normalize dates to ISO 8601 format (YYYY-MM-DD)	normalize_date
Normalize street names across languages	normalize_street
Normalize text for matching	normalize_text
Tokenize numeric fields, expanding ranges into individual numbers	numeric_tokens
Plan a Search Strategy from Raw Inputs	plan_strategy
Prepare Data for Record Linkage Search	prepare_search_data
Read the Token Rarity Distribution	rarity_distribution
Bar chart of median token rarity per column	rarity_histogram
Recommendations from a Diagnostic Object	recommendations
Group Matched Pairs into Entities	resolve_entities
Sample Matches for Review	sample_matches
Kernel density of the score distribution	score_density
Score Embedding Pairs Using Cosine Similarity	score_embeddings
Bar chart of the pre-binned score distribution	score_histogram
Search for Candidate Matches Between Tables	search_candidates
Define a Search Strategy for Record Linkage	Search_Strategy search_strategy
Histogram of sampled pairwise cosine similarities	similarity_histogram
Configure rIP smoothing for a search strategy	smooth_rip smooth_rip_identity smooth_rip_log smooth_rip_offset smooth_rip_softmax
Line plot of cumulative base coverage by stage	stage_coverage_plot
Grouped bar chart of score distributions by stage	stage_score_plot
Multilingual Street-Name Stopwords	street_stopwords
Multilingual Street-Type Normalization Dictionary	street_types
Strip vowels from text (consonant skeleton)	strip_vowels
Summarise a Match Result	summarise_matches
Target dataset for record linkage example	target_example
Horizontal bar chart of per-token score contributions, coloured by column	token_contribution_plot
Bar chart of average tokens per record per column	token_frequency_plot
Convert tokens to shape signatures	token_shapes
Bar chart of top-1 vs top-2 score gap distribution (candidates only)	top_gap_density
Map tokens to canonical groups with a lookup table	use_dictionary
Bar chart of vocabulary overlap between base and target per column	vocab_overlap_plot
Split text into word tokens	word_tokens
Workshop external directory (target) for record linkage examples	workshop_listings
Multi-year workshop panel for cross-year linkage examples	workshop_panel
Workshop guild register (base) for record linkage examples	workshop_register

Package: joinery 1.0.1

joinery: Heuristic Index-Based Record Linkage

Citation

Readme and manuals

Help Manual

Usage by other packages (reverse dependencies)