... differences hold to alesser degree when a partial dictionary is provided.With MLHMM, different tokens of the same wordtype are usually assigned to the same cluster, buttypes are assigned to clusters ... We would also like to thankNoah Smith for providing us with his data sets.Eisner, 2005). Nearly all of these approaches haveone aspect in common: the goal of learning is to identify the set ... tagdictionary to no dictionary at all. We introduce theuse of a new information-theoretic criterion, varia-tion of information (Meilˇa, 2002), which can be used to compare a gold standard clustering to...