#21 · Primary category: NLP Tools & Text Processing

libpostal

address address-parser c deduping deduplication international machine-learning natural-language-processing nlp record-linkage

A C library for parsing/normalizing street addresses around the world. Powered by statistical NLP and open geo data.

Project last updated:05/13/26

GitHub Stars

4.9K

Forks

470

Contributors

83

License

MIT

Why we included this project

Most geocoding stacks that handle global addresses quietly depend on libpostal somewhere in the chain. It is a C library that reads free-form street addresses from dozens of countries and splits them into house number, street, city, and postcode, then normalizes the result so the same physical place written several ways comes out as one consistent string. That step matters before you deduplicate records or join addresses across databases, and the project backs popular Python and Go bindings plus higher-level geocoding tools. It is not a standalone application or a hosted API, so you embed it in your own pipeline. The payoff is offline, permissive-licensed address processing that runs locally and never sends sensitive location data to a third party.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category