基于多特征的语义角色标注一致性计算方法研究  

A Method for Semantic Roles Labeling Consistency Calculation Based on Multi-features

作　　者：柯永红[1] 朱永福 穗志方[2] 俞士汶[2] KE Yonghong1, ZHU Yongfu2, SUI Zhifang2, YU Shiwen2(1. School of Chinese Language and Literature, Beijing Normal University, Beijing 100871; 2. Institute of Computational Linguistics, Peking University, Beijing 10087)

机构地区：[1]北京师范大学文学院,北京100871 [2]北京大学计算语言学教育部重点实验室,北京100871

出　　处：《北京大学学报（自然科学版）》2018年第3期475-480,共6页Acta Scientiarum Naturalium Universitatis Pekinensis

基　　金：国家重点基础研究发展计划(2014CB340504)资助

摘　　要：综合语义角色标注语料的格式、标签结构、标签内容和用户可信度等多个特征,实现一种自动的语义角色标注一致性计算方法。实验证明,该方法对错误标注的召回率高,并且速度快,结果稳定,可以大幅度地提高语义角色标注一致性检查的效率。The authors state an automatic method for semantic role labeling consistency calculation, based on the features of annotated corpus＇ format, structure, content and user performances. The expriment shows that the proposed method is fast, stable and has high recall rate, and it can greatly improve the quality and efficiency.

关 键 词：语料库 多特征 一致性计算 语义角色标注 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]