Stemmer for Serbian language, built on Croatian stemmer from Natural Language Processing group of Faculty of Humanities and Social Sciences, University of Zagreb.
Stemmer works for Serbian Latin Alphabet, replaces special characters (č -> c, ć -> c, dž -> dz, đ -> dj, š -> s, ž -> z), and returns lowercase characters.
In general, stemmer works best on adjectives and nouns.
Download rules.txt, transformations.txt and serbian_stemmer.py files and put them in same directory.
from serbian_stemmer import stem
output_text = stem(input_text)
Stemmer is written in Python 3.4.