id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_37wft6eehnazxaxgenjumsuamm	Sujith Ravi	Parallel Algorithms for Unsupervised Tagging	2014	13	.pdf	application/pdf	8035	678	66	and multi-step heuristics for model minimization, our approach is a simple greedy approximation algorithm DMLC (DISTRIBUTEDMINIMUM-LABEL-COVER) that solves this sequence labeling tasks: Part-of-Speech tagging for multiple languages (including lowresource languages), with complete and incomplete dictionaries, and supertagging, a In this work, we tackle the problem of unsupervised sequence labeling using tag dictionaries. Ravi and Knight (2009) explores the idea of performing model minimization followed by EM training to learn taggers. large data and grammar sizes, and does not require the corpus or label set to fit into memory. A more popular approach is to learn from POS-tag dictionaries (Merialdo, 1994; Ravi and Knight, 2009), incomplete dictionaries (Hasan and Ng, 2009; Garrette and word wij we associate a set of possible tags Tij. We from the raw and test data to perform model minimization followed by unsupervised EM training. raw training data along with the word-tag dictionary	./cache/work_37wft6eehnazxaxgenjumsuamm.pdf	./txt/work_37wft6eehnazxaxgenjumsuamm.txt