基于规则的依存树库错误自动检测与分析.PDFVIP

北京大学学报(自然科学版) 第 52 卷 第 1 期 2016 年 1 月 Acta Scientiarum Naturalium Universitatis Pekinensis, Vol. 52, No. 1 (Jan. 2016) doi: 10.13209/j.0479-8023.2016.005 基于规则的依存树库错误自动检测与分析 † 史林林 邱立坤 亢世勇 鲁东大学文学院 , 烟台 264025; † 通信作者 , E-mail: qiulikun@pku.edu.cn 摘要 尝试将依存树转化为短语结构树, 并基于规则的方法自动检测出人工标注结果中的错误。将该方法应 用于已经过两遍人工校对的北京大学多视图依存树库 , 从 50275 个句法树中发现 1529 处错误 , 正确率为 100%。进一步, 所有错误可以分为 3 个层次 : 分词错误、词性与句法角色不符、句法角色错标。该方法可以 有效提高依存树库的质量, 并且适用于各类型的依存树库。 关键词 树库 ; 词性 ; 句法角色; 错误检测 中图分类号 TP391 Rule-Based Detection and Analysis of Annotation Errors in Dependency Treebank † SHI Linlin, QIU Likun , KANG Shiyong School of Chinese Language and Literature, Ludong University, Yantai 264025; † Corresponding author, E-mail: qiulikun@pku.edu.cn Abstract The authors try to transform dependency tree into phrase structure tree, and detect annotation errors automatically based on manual rules. The method is used in processing Peking University Multi-view Chinese Treebank (PMT). Although PMT has been manually checked twice before processed by this method, 1529 errors are detected among the 50275 sentences and the precision is 100%. The errors mainly belong to three types: word segmentation error, mismatching between POS and syntactic role, and syntactic role error. This method can further improve treebank quality, and be applied to other dependency treebanks. Key words treebank; part of speech; syntactic role; error detection 树库是在分词和词性标注的基础上 , 对句子中 注人员偶然误操作导致的。本文把经过两遍人工校