# dedupeio/dedupe

:id: A python library for accurate and scalable fuzzy matching, record deduplication and entity-resolution.

Repository: https://github.com/dedupeio/dedupe
Canonical: https://ross.abutalabs.com/products/dedupe
Homepage: https://docs.dedupe.io
Language: Python
License: MIT
License Family: permissive
Topics: dedupe, record-linkage, python, python-library, entity-resolution, dedupe-library, de-duplicating, datamade, clustering
Last push: 2025-07-29T00:18:20+00:00

## Health v2 (maintenance only)
Score: 48/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 34, release rhythm 35, longevity 100
- inputs: {"age_days": 5248, "days_push": 401, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4507, forks 576 (observed 2026-08-28T04:08:52.516919+00:00)

## What it is
Dedupe is a Python library that uses machine learning to perform fuzzy matching, deduplication, and entity resolution on structured data. It learns from human-labeled training examples to automatically derive optimal matching weights and blocking rules, scaling to large databases.

## Use cases
- remove duplicate entries from a spreadsheet of names and addresses
- link a customer list to an order history list without shared unique IDs
- identify which campaign contribution records were made by the same person despite name variations
- deduplicate messy CSV files
- perform entity resolution across large structured datasets
- train a custom fuzzy matching model from human-labeled examples

## When to choose
- you need to deduplicate or link structured records with fuzzy, inconsistent text fields
- you have (or can collect) some human-labeled training pairs and want the model to learn matching rules automatically
- you want a well-established, MIT-licensed Python library that runs on a laptop without special infrastructure

## When to avoid
- you need exact matching on clean keys or unique identifiers - simple database joins suffice
- you need real-time streaming deduplication rather than batch processing
- you want a turnkey GUI or hosted service rather than writing Python code (consider Dedupe.io instead)

## Facets
- artifact type: library
- maturity: stable
- function: machine-learning, nlp, data-science, etl
- domain: data-science, machine-learning, developer-tools
- platform: python, cross-platform
- tags: fuzzy-matching, record-linkage, entity-resolution, deduplication, active-learning, clustering, data-engineering

## Member repositories
- dedupeio/dedupe (main) score 48

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:52.516919+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:20:22.374576+00:00, confidence not recorded.
  - readme: https://github.com/dedupeio/dedupe (fetched 2026-08-28T04:08:52.516919+00:00, sha c2ad4c18ce62)
  - homepage: https://docs.dedupe.io (fetched 2026-08-29T09:06:31.048938+00:00, sha 47cc13cab31a)
  - site_page: https://docs.dedupe.io/en/latest/API-documentation.html (fetched 2026-08-29T09:06:31.060121+00:00, sha 7ea632584033)
  - site_page: https://docs.dedupe.io/ (fetched 2026-08-29T09:06:31.065106+00:00, sha 47cc13cab31a)
  - site_page: https://dedupe.io/pricing (fetched 2026-08-29T09:06:31.058257+00:00, sha 15ff7c392491)
- Data as of 2026-08-30T08:39:29.467469+00:00.