面向药物研发的大规模数据语义整合与挖掘模式探索  

Research on Big Data Semantic Integration and Mining Pattern for Drug Discovery

作　　者：钱庆[1] 洪娜[1] 李姣[1] 

机构地区：[1]中国医学科学院医学信息研究所

出　　处：《数字图书馆论坛》2014年第3期19-25,共7页Digital Library Forum

基　　金：国家“十二五”科技支撑计划项目课题“科技知识组织体系的协同工作系统和辅助工具开发”(编号:2011BAH10B02);国家社会科学基金项目“关联数据中潜在知识关联的发现方法研究”(编号:11CTQ016)的研究成果之一

摘　　要：整合并在语义层面上充分互连药物研发数据，将有利于从全局、系统化的视角开展药物研发工作，同时也有助于预测药物的不良副作用、加快药物研发流程、缩减药物开发成本等。文章试图探索语义技术如何支持药物研发数据的整合和挖掘，通过基于知识组织体系的语义标注，以及多类型实体互连策略构建充分互连的药物关联数据，支持药物研究人员对这些大量复杂实体及其关系的查找、探索和知识发现，从而帮助药物研发人员和，I盘床工作者更好地利用大规模药物数据，解决药物研发面临的实际问题。Integrating and linking drug discovery data in semantic level is helpful to make drug discovery research implement on global, systematic, and predicable perspectives, also it will accelerate the research flow and reduce the cost of development. In this paper, we explore how semantic technology supports the integrating and mining of drug data, through semantic annotation based on knowledge organization system and multi-type entity linking strategy to construct interlinked data for drug discovery. Our research supports the search, exploration and mining of large scale entities and complex entity relations. The object of our effort is to make the most of drug big data for supporting researchers and clinicians to resolve the real problems existing in drug discovery.

关 键 词：语义整合 实体互连 RDF 可视化 药物研发 

分 类 号：G254[文化科学—图书馆学]