Yu-Ming Hsieh

Chinese treebanks and grammar extraction

Preparation of knowledge bank is a very difficult task. In this paper, we discuss the knowledge extraction from the manually examined Sinica Treebank. Categorical information, word-to-word relation, word collocations, new syntactic patterns and sentence structures are obtained. A searching system for Chinese sentence structure was developed in this study. By using pre-extracted data and SQL commands, the system replies the user’s queries efficiently. We also analyze the extracted grammars to study the tradeoffs between the granularity of the grammar rules and their coverage as well as ambiguities. It provides the information of knowing how large a treebank is sufficient for the purpose of grammar extraction. Finally, we also analyze the tradeoffs between grammar coverage and ambiguity by parsing results from the grammar rules of different granularity.

Linguistically-motivated grammar extraction, generalization and adaptation

In order to obtain a high precision and high coverage grammar, we proposed a model to measure grammar coverage and designed a PCFG parser to measure efficiency of the grammar. To generalize grammars, a grammar binarization method was proposed to increase the coverage of a probabilistic context-free grammar. In the mean time linguistically-motivated feature constraints were added into grammar rules to maintain precision of the grammar. The generalized grammar increases grammar coverage from 93% to 99% and bracketing F-score from 87% to 91% in parsing Chinese sentences. To cope with error propagations due to word segmentation and part-of-speech tagging errors, we also proposed a grammar blending method to adapt to such errors. The blended grammar can reduce about 20~30% of parsing errors due to error assignment of pos made by a word segmentation system.

Ambiguity Resolution for Vt-N Structures in Chinese

The syntactic ambiguity of a transitive verb (Vt) followed by a noun (N) has long been a problem in Chinese parsing. In this paper, we propose a classifier to resolve the ambiguity of Vt-N structures. The design of the classifier is based on three important guidelines, namely, adopting linguistically motivated features, using all available resources, and easy integration into a parsing model. The linguistically motivated features include semantic relations, context, and morphological structures; and the available resources are treebank, thesaurus, affix database, and large corpora. We also propose two learning approaches that resolve the problem of data sparseness by autoparsing and extracting relative knowledge from large-scale unlabeled data. Our experiment results show that the Vt-N classifier outperforms the current PCFG parser. Furthermore, it can be easily and effectively integrated into the PCFG parser and general statistical parsing models. Evaluation of the learning approaches indicates that world knowledge facilitates Vt-N disambiguation through data selection and error correction.

Improving Chinese Parsing with Special-Case Probability Re-estimation

Syntactic patterns which are hard to be expressed by binary dependent relations need special treatments, since structure evaluations of such constructions are different from general parsing framework. Moreover, these different syntactic patterns (special cases) should be handled with distinct estimated model other than the general one. In this paper, we present a special-case probability re-estimation model (SCM), integrating the general model with an adoptable estimated model in special cases. The SCM model can estimate evaluation scores in specific syntactic constructions more accurately, and is able for adopting different features in different cases. Experiment results show that our proposed model has better performance than the state-of-the-art parser in Chinese.