基于最大熵的汉语篇章结构自动分析方法 - 模式识别国家重点实验室.pdf

网络出版地址：/kcms/detail/11.2442.N0926.001.html 北京大学学报 ( 自然科学版) Acta Scientiarum Naturalium Universitatis Pekinensis doi: 10.13209/j.0479-8023.2014.021 基于最大熵的汉语篇章结构自动分析方法 † 涂眉 周玉 宗成庆 中国科学院自动化研究所模式识别国家重点实验室, 北京 100190; † 通信作者 , E-mail: mtu@ 摘要 在标有复句逻辑语义关系的清华汉语树库上, 研究汉语篇章语义片段自动切分以及篇章关系的自动标 注方法。比较了不同序列标注模型对汉语篇章语义单元切分的性能, 并提出基于最大熵模型的汉语篇章结构 分析方法。实验结果表明, 篇章语义单元自动切分的 F 值能达到 89.1%, 当篇章语义结构树的高度不超过 6 层时, 篇章语义关系标注的 F 值为 63% 。 关键词 语义片段自动切分 ; 篇章结构分析 ; 逻辑语义关系 ; 树库 中图分类号 TP391 Automatically Parsing Chinese Discourse Based on Maximum Entropy † TU Mei , ZHOU Yu, ZONG Chengqing National Laboratory of Pattern Recongnition (NLPR), Institute of Automation, Chinese Academy of Sciences, Beijing 100190; † orresponding author, E-mail: mtu@ Abstract The authors focus on how to segment semantic units in Chinese discourse and how to label relations among semantic units automatically. During the parsing process, several sequence labelling methods are compared for discourse segmentation, while a maximum entropy-based training and decoding algorithm is specially proposed.Experiments are done based on Tsinghua Chinese Treebank which is annotated with logical and semantic relations at complex-sentence level. Experimental results show that F-score of discourse