【作者】 熊德意； 刘群；

【Author】 Devi Xiong Qun Liu(Institute of Computing technology, the Chinese Academy of Sciences, Beiiing, 100080); ~2(Graduate School of the Chinese Academy of Sciences. Beijing, 100039)

【机构】 中国科学院计算技术研究所；

【摘要】 我们的汉语统计句法分析模型从3个方面融合丰富的语言特征知识:1)利用非递归名词短语界的相对确定性重新标注树库中的名词短语;2)设计新的中心词映射表;3)引进上下文配置框架。这些语言特征知识使模型的性能提高了10%。更多还原

【Abstract】 Rich linguistic features are incorporated into our model for Chinese statistical parsing by the following three ways. First of all, non-recursive noun phrases are annotated in the Penn Chinese Treebank because of their strong mark of boundaries. Second, a new head percolation table is designed based on Xia’s table. The last linguistic feature our model uses is context configuration flame which builds a platform for incorporating knowledge about commas, coordination constructions and so on. All these three linguistic features give about 10% improvement of our model.更多还原