基于知识元的文献挖掘研究——以粤海关文献资料为例

发布时间：2018-10-26 11:40

【摘要】：【目的/意义】文献挖掘是文本挖掘领域中的重要研究方向,文献挖掘技术在信息化时代发挥着越来越重要的作用。【方法/过程】首先在文献挖掘过程中引入知识元概念,提出了一套基于知识元的文献挖掘处理模式。其次针对海关贸易档案文献资料,提出了一种五元组结构的知识元表示方法,并用领域知识元本体进行领域知识元间的组合链接。同时针对期刊文献资料,提出了五类知识元共同表示一篇期刊文献的知识元表示方法。最后以"粤海关"相关贸易档案文献资料和中文期刊文献资料为例,运用所提方法进行知识元的抽取和表示,在知识元的基础上进行知识元间的组合链接进而实现快速精确的领域知识挖掘。【结果/结论】基于知识元的文献挖掘模型,能够增强文献挖掘过程的可重复操作性并提高挖掘成果的可重复利用性。
[Abstract]:[purpose / significance] document mining is an important research direction in the field of text mining. Literature mining technology plays a more and more important role in the information age. [method / process] firstly, the concept of knowledge element is introduced in the process of document mining. A set of document mining processing model based on knowledge element is proposed. Secondly, a knowledge element representation method based on the five-tuple structure is proposed, and the domain knowledge element ontology is used to link the domain knowledge elements. At the same time, five kinds of knowledge elements are proposed to represent a journal document. Finally, taking "Guangdong Customs" related trade archives and Chinese periodical literature as an example, the proposed method is used to extract and express the knowledge elements. On the basis of the knowledge element, the combination link between the knowledge elements is carried out to realize the fast and accurate domain knowledge mining. [results / conclusion] the document mining model based on the knowledge element, It can enhance the repeatability of the document mining process and improve the reusability of the mining results.
【作者单位】：大连理工大学系统工程研究所;软件架构国家重点实验室(东软集团股份有限公司);大连理工大学城市学院;
【基金】：国家自然科学基金资助项目(71171030) 新世纪优秀人才支持计划资助(NCET-11-0050) 文化产业发展专项资金资助项目(教财司预函[2012]343号) 软件架构国家重点实验室开放课题(SKLSAOP1701)
【分类号】：G254

【相似文献】