Comparing Multiple Dictionaries to Improve Part-of-Speech Tagging of Ainu Language

† Department of Computer Science, Kitami Institute of Technology {ptaszynski,f-masui}@cs.kitami-it.ac.jp ‡ Independent Researcher § Professor Emeritus of Hokkai Gakuen University karol nowakowski@interia.pl Abstract In this paper we present our research in improving POST-AL, a POS tagger for Ainu language. We focused on improving its dictionary base by comparing various Ainu language resources. We discuss the differences between the dictionaries and evaluate the system when each dictionary is applied. The experiments indicate that the size of the dictionary is less important than the way it was created. Moreover, dictionary records should be refined and unified.