摘要
提出一种基于案例分析的文本数据抽取方法,通过将知识进行特征化表示,借助"用户特征—案例特征—案例知识"三者之间的映射关系和概念模块间的知识关联,完成复杂信息的知识抽取,同时引入增量式案例知识学习模型,有效地避免了因人工干预导致的知识拓展的不连续性,提高了抽取过程的识别效率.
In this paper, we propose an information extraction technology based on case study, which realizes the knowledge extraction of complex information by means of characteristically expressing knowledge and via the help of the mapping relations among ＂the user characteristic-case knowledge-case characteristic＂ and the associations of the knowledge module. Meantime, the bootstrapping knowledge learning model is introduced to effectively avoid the discontinuousness of knowledge expansion caused by human interference~ thus, the recognition efficiency is greatly enhanced.
出处
《辽宁师专学报（自然科学版）》
2014年第2期1-3,65,共4页
Journal of Liaoning Normal College(Natural Science Edition)
关键词
文本信息
信息抽取技术
知识抽取
text information
information extraction technology
knowledge extraction
