Treetagger

TreeTagger is a widely used tool for annotating text with linguistic information, primarily part-of-speech tags and lemmas, along with basic morphological features. It was developed by Helmut Schmid at the University of Stuttgart and has been available since the late 1990s. The system is designed to be language independent in principle and ships with parameter files for a variety of languages, trained on language-specific corpora to produce consistent annotations.

The tool operates as a standalone program with a simple command-line interface. Users input plain text and receive tagged output that typically includes the surface form, the assigned POS tag, the lemma, and optional morphological attributes. Outputs can be formatted for easy integration into downstream processing pipelines or corpusAnnotation workflows.

TreeTagger relies on language-specific parameter files created through supervised training, enabling tagging and lemmatization without reimplementing language rules for each new corpus. Adding a new language requires appropriate annotated data and training of a new parameter file. The system is designed to be extensible and portable, with wrappers and interfaces available for several programming languages, including Python, Java, and Perl, facilitating integration into NLP projects.

Used in academia and industry alike, TreeTagger has been employed in corpus annotation, linguistic research, and educational settings. It remains a reference tool for multilingual POS tagging and lemmatization, valued for its simplicity, broad language coverage, and reproducible outputs.