摘要
本文主要介绍了一个利用最大熵进行实体名识别的系统以及所采用的模型和选取的特征。这些特征包括单词本身的词法词态特征和上下文信息。利用这些在任何语言的文本上都极易获得的特征 ,我们采用最大熵分类器构建了一个基准系统。在此基础上 ,我们首先通过网络资源建立了实体名词典知识库 ;并利用词典和基准系统在未标注语料上抽取出现的实体名作为辅助的训练语料 ;最后再将这些语料加入训练。实验结果表明 ,辅助的训练语料能够在一定程度上提高系统的性能。
In this paper, we describe a system that applies maximum entropy (ME) models to the task of named entity recognition (NER). We start with an annotated corpus and a set of features. These features include the morphological features and context information. They are easily obtainable for almost any language. We build a baseline NE recognizer based on these features. We first construct a named entity thesaurus based on the information on the web. Then the baseline together with the thesaurus is used to extract the named entities and their context information from additional un annotated data. In turn, these data are incorporated into the final recognizer. The experiments prove that these data could further improve the recognition accuracy.
出处
《中文信息学报》
CSCD
北大核心
2005年第2期7-11,27,共6页
Journal of Chinese Information Processing
基金
国家自然科学基金资助项目 (6 0 10 30 14 )
上海市科委重点研究项目资助 (0 35 115 0 2 8)
关键词
计算机应用
中文信息处理
实体名识别
最大熵
未标注语料
computer application
Chinese information processing
named entity recognition
maximum entropy
un-annotated data
