Stemmers are available here under the GNU GPL license.
The stemmer uses morphological rules to generate all possible base forms. Most of them (80% in case of Slovenian) don't exist in the real language. But if we have a dictionary, we can filter out the false positives.
Please send your complaints and suggestions to aprimc@gmail.com.
Slovenian stemmer generates most forms of productive word classes.
TODO: comparation of adjectives, adjectives with definite form only, some irregular verbs and nouns, participles, pronouns, adverbs
Slovenian stemmer generates most forms of productive word classes.
TODO: comparation of adjectives, missing classes of nouns (-lac [m.], -vođa [m.], -in [m.]...), some irregular verbs and nouns, participles, pronouns, adverbs
TODO: all