Contingency and Comparison Relation Labeling and Structure Prediction in Chinese Sentences
نویسندگان
چکیده
Unlike in English, the sentence boundaries in Chinese are fuzzy and not well-defined. As a result, Chinese sentences tend to be long and consist of complex discourse relations. In this paper, we focus on two important relations, Contingency and Comparison, which occur often inside a sentence. We construct a moderate-sized corpus for the investigation of intrasentential relations and propose models to label the relation structure. A learning based model is evaluated with various features. Experimental results show our model achieves accuracies of 81.63% in the task of relation labeling and 74.8% in the task of relation structure prediction.
منابع مشابه
Terminology of Combining the Sentences of Farsi Language with the Viterbi Algorithm and BI-GRAM Labeling
This paper, based on the Viterbi algorithm, selects the most likely combination of different wording from a variety of scenarios. In this regard, the Bi-gram and Unigram tags of each word, based on the letters forming the words, as well as the bigram and unigram labels After the breakdown into the composition or moment of transition from the decomposition to the combination obtained from th...
متن کاملSemantic Labeling of Chinese Serial Verb Sentences Based on Feature Structure
Parsing Chinese serial verb sentences is a key issue in NLP. Many controversies arise from serial verb sentences. This paper puts forward a novel model “the Feature Structure theory” to resolve the semantic labeling of Chinese serial verb sentences. We analyze the difficulties in annotating these sentences, and compare Feature Structure with traditional dependency structure. Feature Structure r...
متن کاملSemantic relation identification for consecutive predicative constituents in Chinese
In this paper, we propose a general methodology for designing semantic role/ relation system. Based on this methodology, we establish a succinct semantic relation system for consecutive predicative constituents for Chinese, which includes serial verb construction, discourse construction, and other constructions describing serial events. This semantic relation system has 13 middle-level classes ...
متن کاملA Hybrid Method for Chinese Entity Relation Extraction
Entity relation extraction is an important task for information extraction, which refers to extracting the relation between two entities from input text. Previous researches usually converted this problem to a sequence labeling problem and used statistical models such as conditional random field model to solve it. This kind of method needs a large, high-quality training dataset. So it has two m...
متن کاملA Comparative Study of the Metaphysical Basis of Ancient Iran-China Political Approach
Considering the cosmology with mythological form of consciousness era as the primary base of metaphysical form and the basis of development toward an integrated cosmology, political ideas has been placed in an organic link with a metaphysical system in the ancient Persians as well as Chinese political thought. Based on considerable similarities among cosmological systems in civilizations e.g....
متن کامل