Semi-supervised Chinese Word Segmentation based on Bilingual Information

Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, pages 1207–1216, Lisbon, Portugal, 17-21 September 2015. c⃝2015 Association for Computational Linguistics. Semi-supervised Chinese Word Segmentation based on Bilingual Information Wei Chen Institute of Automation, Chinese Academy of Sciences, Beijing, 100190, China wei.chen.media@ia.ac.cn Bo Xu Institute of Automation, Chinese Academy of Sciences, Beijing, 100190, China xubo@ia.ac.cn Abstract This paper presents a bilingual semi- supervised Chinese word segmentation (CWS) method that leverages the nat- ural segmenting information of English sentences. The proposed method in- volves learning three levels of features, namely, character-level, phrase-level and sentence-level, provided by multiple sub- models. We use a sub-model of condi- tional random fields (CRF) to learn mono- lingual grammars, a sub-model based on character-based alignment to obtain ex- plicit segmenting knowledge, and anoth- er sub-model based on transliteration sim- ilarity to detect out-of-vocabulary (OOV) words. Moreover, we propose a sub-model leveraging neural network to ensure the proper treatment of the semantic gap and a phrase-based translation sub-model to s- core the translation probability of the Chi- nese segmentation and its corresponding English sentences. A cascaded log-linear model is employed to combine these fea- tures to segment bilingual unlabeled data, the results of which are used to justify the original supervised CWS model. The eval- uation shows that our method results in su- perior results compared with those of the state-of-the-art monolingual and bilingual semi-supervised models that have been re- ported in the literature. 1 Introduction Chinese word segmentation (CWS) is generally accepted to be a necessary first step in most Chi- nese NLP tasks because Chinese sentences are written in continuous sequences of characters with no explicit delimiters (e.g., the spaces in English). Many studies have been conducted in this area, re- sulting in extensive investigation of the problem of CWS using machine learning techniques in recent years. However, the reliability of CWS that can be achieved using machine learning techniques re- lies heavily on the availability of a large amount of high-quality, manually segmented data. Because hand-labeling individual words and word bound- aries is very difficult (Jiao et al., 2006), producing segmented Chinese texts is very time-consuming and expensive. Although a number of manual- ly segmented datasets have been constructed by various organizations, it is not feasible to com- bine them into a single complete dataset because of their incompatibility due to the use of various segmenting standards. Thus, it is difficult to build a large-scale manually segmented corpus, and the resulting lack of such a corpus is detrimental to further enhancement of the accuracy of CWS. To address the scarcity of manually segment- ed corpora, a number of semi-supervised CWS approaches have been intensively investigated in recent years. These approaches attempt to ei- ther learn the predicted label distribution (Jiao et al., 2006) or extract mutual information ((Liang et al., 2005); (Sun and Xu, 2011); (Zeng et al., 2013a)) from large-scale monolingual unlabeled data to update the baseline model (from manual- ly segmented corpora). In addition to these tech- niques, several …

Content truncated to 500 words