CombiTagger: A System for Developing Combined Taggers
Verena Henrich, Timo Reuter, Hrafn Loftsson · 2009
The main task of part-of-speech (PoS) tagging is to as-sign the appropriate morphosyntactic category to each word in a sentence. A combination of different PoS tag-gers usually results in higher tagging accuracy than ob-tained by the use of only a single tagger. We present a new language and tagset independent system, Combi-Tagger, which combines automatically the output of several taggers. The system, which is open source, pro-vides algorithms for simple and weighted voting, but it is extensible so that other combination algorithms can be added easily. We demonstrate the functionality of CombiTagger by using it to develop and evaluate com-bined taggers for Icelandic. The most accurate indi-vidual tagger obtains an accuracy of 91.83%. Combi-Tagger achieves 93.09%-93.41 % accuracy by combin-ing the output of five or six taggers using simple and weighted voting.