【摘要】 该文提出了一个统计语义分析器 ,它能够发现中文句子中的语义依存关系 .这些语义依存关系可以用于表示句子的意义和结构 .语义分析器在 1百万词的标有语义依存关系的语料库 (语义依存网络语料库 ,SDN)上训练并测试 ,文中设计、实现了多个实验以分析语义分析器的性能 .实验结果表明 ,分析器在非限定领域中表现出了较好的性能 ,分析正确率与中文句法分析器基本相当 .更多还原

【Abstract】 This paper presents a statistical semantic parser, which could discover the semantic dependency relations in a Chinese sentence. By these relations the meaning and structure of sentence could be represented. The parser was trained and evaluated on a 1M-word-scale corpus annotated with semantic dependency, Semantic Dependency Net (SDN). Various experiments were carried out to analyze the performance of the parser, and good performance was shown in parsing domain-unrestricted text. The accuracy reached as high level as the Chinese syntactic parser.更多还原