Joint Chinese Word Segmentation and POS Tagging Using an Error-Driven Word-Character Hybrid Model

摘要

In this paper, we present a discriminative word-character hybrid model for joint Chinese word segmentation and POS tagging. Our word-character hybrid model offers high performance since it can handle both known and unknown words. We describe our strategies that yield good balance for learning the charncteristics of known and unknown words and propose an error-driver peliey that delivers such balance by acquiring examples of unknown wo ds from particular errors in a training corpus. We describe an efficient fiamework for training our model based on the Margin Infused Relaxed Algorilhm (MIRA). evaluate our approach on the Penn Chinese Treebank. and show that it achieves superior performance compared to the state-of-the-art approaches reported in the literature.