omorfi

Omorfi is an open-source Turkish morphological analyzer and generator used in natural language processing. It provides morphological analysis, lemmatization, part-of-speech tagging, and surface-form generation for Turkish words. The system relies on a large lexicon of Turkish lemmas and a set of finite-state rules that capture inflection, derivation, and compounding patterns in Turkish grammar. Omorfi is designed to be integrated into NLP pipelines and can output analyses in machine-readable formats for downstream tasks such as parsing, tagging, and translation.

Developed by researchers in Turkey and contributors from the NLP community, its development has spanned the 2000s and 2010s. It has been used in academic research and as a component in open-source NLP stacks, with emphasis on coverage of Turkish morphophonology, including vowel harmony and agglutinative morphology.

Key features include morphological analysis for nouns, verbs, and adjectives; support for root extraction; generation of inflected forms from a lemma and grammatical features (case, number, tense, aspect, mood, person, voice); and handling of clitics and several irregular forms. The project provides command-line tools and libraries for integration with Python or Java-based workflows and can be used to produce annotated corpora.

Omorfi is cited in Turkish NLP research and discussed as part of the broader ecosystem of Turkish language resources. See also Turkish language, Turkish morphology, natural language processing, finite-state transducers, and lemmatization.