Splink (Python package)

Open-source Python package for probabilistic linkage and deduplication of records without unique identifiers.

Official site ↗

At a glance

Organization: Splink project
Type: record linkage library
Access: open source
Deployment: application runtime
Lifecycle: active

When to consider it

When an analyst needs inspectable pairwise predictions and record clusters within or across datasets

Considerations

Standardize input columns first. Splink works best with multiple informative attributes and is not designed for a single bag-of-words column. Linkage probabilities and clusters require evaluation; they do not prove identity.

Connections

In areaData & Preparation
Uses approachProbabilistic record linkage
Supports functionFind records for the same entity

Sources

Splink documentation overviewSplink project · checked 2026-10-04
Link type: linking, deduping or bothSplink project · checked 2026-10-04
Evaluating record linkageSplink project · checked 2026-10-04

Last reviewed 2026-10-04