| 基于决策树的现代汉语中任职关系抽取研究 |
| |
| 引用本文： | 帅训波,马书南.基于决策树的现代汉语中任职关系抽取研究[J].昆明理工大学学报(理工版),2009,34(4):27-31. |
| |
| 作者姓名： | 帅训波 马书南 |
| |
| 作者单位： | 1. 中国石油勘探开发研究院,廊坊分院,地球物理与信息研究所,河北,廊坊,065007
2. 北京工业大学,计算机学院,北京,100022 |
| |
| 基金项目： | 河北省科学技术进步成果资助项目 |
| |
| 摘 要： | 在命名实体识别的研究基础之上，论文把抽取人名实体与机构实体间的任职关系看成分类问题．即根据现代汉语句子中任职动词的类别属性将任职关系信息抽取模式分类．应用决策树的方法确定句子的抽取模式，实现人在机构中的任职关系信息抽取．并对建立的基于该决策树的任职关系抽取系统进行开放测试，平均召回率和精确率分别为91．47％和89．15％，实验结果表明，基于决策树的现代汉语中任职关系抽取是一种值得继续探讨的方法．
|
| 关 键 词： | 命名实体识别 决策树 信息抽取 自然语言处理 |
Research of Duty Information Extraction in Chinese Based on Decision Tree |
| |
| SHUAI Xun-bo,MA Shu-nan.Research of Duty Information Extraction in Chinese Based on Decision Tree[J].Journal of Kunming University of Science and Technology(Natural Science Edition),2009,34(4):27-31. |
| |
| Authors: | SHUAI Xun-bo MA Shu-nan |
| |
| Affiliation: | SHUAI Xun-bo MA Shu-nan ( 1. Institute of Geophysics and Information, Langfang Branch of Research Institute of Petroleum Exploration and Development, Langfang, Hebei 065007, China; 2. College of Computer Science, Beijing University of Technology, Beijing 100022, China) |
| |
| Abstract: | Based on the named entity recognition research, duty information extraction is taken as a question of classification first, that is, information extraction mode is classified by duty verb in Chinese sentences. Decision tree is used to select appropriate extraction mode, which solves the problem of duty information extraction. A Chinese duty information extraction system based on the decision tree is thus realized. For an open test, its average recall rate is 91.47% and average precision is 89.15%. Experimental results show that the information extraction method based on decision tree is worth to continue for further research. |
| |
| Keywords: | named entity recognition decision tree information extraction natural language processing |
| 本文献已被 维普 万方数据 等数据库收录！ |
|
