We provide two stemmers for Hungarian: a light stemmer and a heavy stemmer. The light stemmer is less invasive than the heavy stemmer altering fewer words. It attempts to stem nouns and adjectives. The heavy stemmer also stems verbs and numerals.

The package (click here) contains Snowball, C and Java versions of both stemmers, as well as a read me file.

We also provide a stopword list for Hungarian.

If you publish results obtained using this resource, please cite this paper:

• Four Stemmers and a Funeral: Stemming in Hungarian at CLEF 2005, A. Tordai and M. de Rijke. In: C. Peters et al, editors, Proceedings CLEF 2005. [PDF]