Previous approaches to Chinese zero pronoun resolution mainly use syntactic information and probabilistic methods, but the context information is ignored. To make full use of the context and semantic information, we build a context-aware model. We propose a key words extraction strategy and design a classification model by using distributed representations as context feature. To our best knowledge, this is the first work using distributed representations in Chinese zero pronoun resolution. Experimental results show that our approach achieves a better performance than previous supervised methods.
CITATION STYLE
Wu, B., & Zhao, T. (2016). A context-aware model using distributed representations for Chinese zero pronoun resolution. In Communications in Computer and Information Science (Vol. 623, pp. 3–11). Springer Verlag. https://doi.org/10.1007/978-981-10-2053-7_1
Mendeley helps you to discover research relevant for your work.