The clumsy NLP JS library!

natural language processing in the browser

General natural language (tokenizing, stemming (English, Russian, Spanish), classification, inflection, phonetics, tfidf, WordNet, jaro-winkler, Levenshtein distance, Dice's Coefficient) facilities for node.

Verified Loop multi-agent system for rigorous research and development in opencode and Claude Code

The lightweight Khmer line breaker with no dependencies.

A Kotlin rewrite of the Stanford Part-Of-Speech Log-Linear tagger

Turn a Universal Dependencies parse into sentences, clauses, phrases and words — every unit traceable to the exact characters it came from.

General natural language (tokenizing, stemming (English, Russian, Spanish), classification, inflection, phonetics, tfidf, WordNet, jaro-winkler, Levenshtein distance, Dice's Coefficient) facilities for node.

A TypeScript parser for CoNLL-U and CoNLL-UL (Universal Dependencies) format with full type definitions

Using this module you can parse CoNLL-U files as a stream of sentence objects. You can also access the low-level line parser, if you want deeper control.