| 基于ART网络的无指导中文共指消解方法 |
| |
| 引用本文： | 李世奇,赵铁军,陈晨,刘鹏远.基于ART网络的无指导中文共指消解方法[J].高技术通讯,2009,19(9). |
| |
| 作者姓名： | 李世奇 赵铁军 陈晨 刘鹏远 |
| |
| 作者单位： | 哈尔滨工业大学教育部-微软语言语音重点实验室,哈尔滨,150001 |
| |
| 基金项目： | 国家自然科学基金，863计划 |
| |
| 摘 要： | 提出了一种基于自适应谐振理论(ART)网络的无指导中文名词短语共指消解方法.该方法充分利用名词短语自身特征,通过改变网络参数动态调节聚类数量,有效地解决了目前聚类共指消解中输出类别数目难以确定的难题.另外采用了一种基于信息增益率的特征选择方法,减少了区分度较弱特征给聚类所带来的干扰.该方法在保证了识别正确率的前提下,不依赖人工标注语料,可直接应用于跨领域的真实文本.最后在ACE中文语料上进行了相关实验,并取得了较好的结果.
|
| 关 键 词： | 共指消解 无指导学习 自适应谐振理论(ART) 自然语言处理 |
An unsupervised approach based on ART network for coreference resolution of Chinese |
| |
| Li Shiqi,Zhao Tiejun,Chen Chen,Liu Pengyuan.An unsupervised approach based on ART network for coreference resolution of Chinese[J].High Technology Letters,2009,19(9). |
| |
| Authors: | Li Shiqi Zhao Tiejun Chen Chen Liu Pengyuan |
| |
| Affiliation: | Li Shiqi Zhao Tiejun Chen Chen Liu Pengyuan (Harbin Institute of Technology MOE-MS Key Laboratory of Natural Language Processing and Speech,Harbin 150001) |
| |
| Abstract: | This paper proposes a novel unsupervised approach for coreference resolution of Chinese based on adaptive resonance theory(ART)Networks .Through making full use of the characteristics of noun phrases and dynamically adjusting the pa- rameters of the networks,the approach can solve the problem in the present clustering coreference resolution that the number of the output categories is hard to determine .Additionally,the approach performs a feature selection process based on the gain ratio criterion to reduce... |
| |
| Keywords: | coreference resolution unsupervised learning adaptive resonance theory (ART) natural language processing |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
