Context-rule Model for Pos Tagging

Modeling these events improves the perplexity to 22.5, a 14% im- provement over the word-based trigram backoff model, and reduces the POS error rate by

In particular, we will asses the effect of different methods for preprocessing (spelling normalisa- tion, as well as word-for-word translation of the text) on the accuracy of

For example, in the CoNLL 2017 shared task on UD parsing (Zeman et al., 2017), UDPipe 1.2 obtained 0.1+% higher average tokenization and word segmentation scores and 0.2% higher

Methods for training with a small labeled set have also been developed (Søgaard, 2011; Garrette and Baldridge, 2013; Duong et al., 2014), but there have not been studies on the

We proposed a joint model of word segmentation, POS tagging and normalization, in which the three tasks benefit from each other. The model is trained on standard corpora, hence there

Inspired by robust generalization and adver- sarial learning we describe a novel approach to learning structured perceptrons for part-of- speech (POS) tagging that is less sensitive

We have demonstrated this claim by drastically reducing the amount of labeled data required for state-of-the-art POS tagging accuracy with word representations derived under the