摘要：

面向非结构化信息的知识发现问题，给出并实现一个基于非结构化信息管理架构(UIMA)的知识发现框架。该框架集成UIMA非结构化信息分析管道，提供将海量、异构非结构化信息抽取为形式化知识的开放式、可扩展的知识发现基础平台和公共设施，构建知识开发协同环境，能实现知识推理辅助下的非结构化信息分析。基于该框架和OWL本体语言，实现了一个针对流感诊断的医疗知识发现系统原型。

关键词: 非结构化信息管理架构, 知识发现, 本体

Abstract: According to the knowledge discovery of the unstructed information, this paper provides a framework of UIMA-based on knowledge discovery. This framewok intergrates a UIMA analysis pipeline and offers an open and extensive conceptual foundation and infrastructure of the knowledge discovery, in order to extract informal knowledge from the massive unstructed information. It also constructs a knowledge environment of cooperation to implement the analysis of unstructed data with the help of reasoning on knowledge. Based on the framework and OWL, it implements a knowledge discovery prototype about medical system.

Key words: Unstructured Information Management Architecture(UIMA), knowledge discovery, ontology

中图分类号: 

• TP391